<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Philips on Medium]]></title>
        <description><![CDATA[Stories by Philips on Medium]]></description>
        <link>https://medium.com/@philips?source=rss-bb923c920c2e------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*xCnLkHFVGmmGMaw6wenMuQ.png</url>
            <title>Stories by Philips on Medium</title>
            <link>https://medium.com/@philips?source=rss-bb923c920c2e------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 13:32:50 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@philips/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Synthetic Data in Healthcare]]></title>
            <link>https://medium.com/philips-technology-blog/synthetic-data-in-healthcare-1a8668648948?source=rss-bb923c920c2e------2</link>
            <guid isPermaLink="false">https://medium.com/p/1a8668648948</guid>
            <dc:creator><![CDATA[Philips]]></dc:creator>
            <pubDate>Fri, 27 Mar 2026 13:07:27 GMT</pubDate>
            <atom:updated>2026-03-27T13:09:09.793Z</atom:updated>
            <content:encoded><![CDATA[<p>Author: <strong>Tim Hulsen</strong>, Senior Data &amp; AI Scientist at Philips &amp; Professor AI &amp; Data Supported Healthcare, Rotterdam University of Applied Sciences.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CjX5fWbJo92vYw7agf37lg.jpeg" /></figure><p><strong>Introduction</strong></p><p>If you look at almost any industry today, artificial intelligence (AI) is tearing down boundaries at breakneck speed. However, in healthcare, we keep hitting a wall. While AI and machine learning applications are rapidly expanding across clinical decision support (CDS), diagnostics, and health system optimization, access to high-quality patient data, the essential fuel of AI, has become a critical bottleneck.</p><p>We are navigating a complex web of legal, ethical, and practical constraints. Regulations like HIPAA in the US, GDPR in Europe, and the newly implemented 2026 EU Artificial Intelligence Act rightly place health AI in the highest-risk category, strictly limiting how we share and reuse real patient records.</p><p>As a Senior Data &amp; AI Scientist at Philips, my team and I grapple with this daily: how do we build life-saving algorithms without compromising patient privacy?</p><p>The use of synthetic data in healthcare has emerged as a promising response to the growing demand for large, diverse, and shareable datasets. In an increasingly data-driven medical ecosystem, synthetic data could allow AI teams to bridge the gap between rapid innovation and bulletproof privacy protection, enabling our developers to safely build and stress-test models while minimizing the risk of patient re-identification.</p><p>Synthetic healthcare data is typically generated using probabilistic or deep learning (DL) models trained on real patient records, including generative adversarial networks (GANs), variational autoencoders (VAEs), and more recently, large language models (LLMs). These approaches aim to reproduce distributions, correlations, and temporal patterns found in real electronic health records, medical images, or clinical text, without directly copying individual patient records. In principle, this enables clinical researchers and developers to work with data that is analytically useful while minimizing the risk of patient re-identification. This privacy-preserving potential is one of the central arguments for synthetic data adoption, particularly in jurisdictions where regulatory compliance significantly restricts data access (El Emam et al., 2023).</p><p><strong>Application examples</strong></p><p>One of the most prominent applications of synthetic data in healthcare lies in the development and training of ML models. Clinical AI systems often require large volumes of labeled data to achieve acceptable performance, yet many healthcare institutions lack sufficient sample sizes or face barriers to data sharing across organizations. Synthetic data can augment existing datasets, enabling earlier experimentation, model prototyping, and stress testing before models are exposed to real patient data. This is particularly relevant in the context of rare diseases and underrepresented demographic groups, where data scarcity can hinder algorithm development and evaluation (Chen et al., 2023). By expanding the apparent size and diversity of datasets, synthetic data can help mitigate some of these limitations, although it does not eliminate the need for real-world validation.</p><p>Examples of large and widely used synthetic healthcare datasets are SyntheticMass and the Data Entrepreneurs’ Synthetic Public Use File (DE-SynPUF) (Gonzales et al., 2023). These datasets consist of synthetic structured data, such as simple tabular files or data in the HL7 (Fast Healthcare Interoperability Resources (FHIR) format. They can be used as demo data in product demonstrations, or even to train AI algorithms with, as they are very similar to real patient data, having relationships and correlations between data fields preserved.</p><p>Medical images, such as magnetic resonance imaging (MRI) images and computed tomography (CT) images, can also be synthesized. At Philips, we are active in Project SEARCH, in which we ensure that synthetic imaging datasets are not only technically accurate but also clinically meaningful and can be trusted by healthcare professionals in areas such as oncology and cardiovascular care. These datasets enable AI models to learn from a diverse and extensive range of population samples, improving their accuracy and robustness, while protecting sensitive patient data and dispelling privacy concerns (Hartjes, 2025).</p><p>Another example of the use of synthetic data at Philips is the Aequitas project (Lemmens et al., 2024), in which Philips works on creating a data synthesizer to repair and mitigate bias, for example, for electrocardiogram (ECG) data.</p><p>Beyond model development, synthetic data plays an important role in software testing, system validation, and education. Health information systems, interoperability frameworks, and CDS tools require extensive testing under realistic conditions, yet using real patient data for this purpose carries legal and ethical risks. Synthetic datasets allow developers to simulate complex clinical scenarios, test edge cases, and evaluate system robustness without exposing sensitive information. In educational settings, synthetic patient records can be used to train clinicians, data scientists, and informaticians, offering realistic learning experiences while avoiding confidentiality concerns (Gonçalves et al., 2020).</p><p><strong>Challenges</strong></p><p>Despite these advantages, the use of synthetic data in healthcare raises significant methodological and ethical challenges. A central concern is data fidelity. Synthetic datasets must preserve clinically meaningful relationships, including correlations between diagnoses, treatments, outcomes, and patient characteristics. If these relationships are distorted or oversimplified, models trained on synthetic data may perform well in artificial settings (during testing and validation) but fail in real clinical environments. Healthcare data is inherently noisy, incomplete, and heterogeneous, and synthetic data that is overly clean or internally consistent may inadvertently misrepresent real-world complexity (Chen et al., 2023).</p><p>Bias propagation represents another critical risk. Synthetic data generation models inevitably reflect the biases present in their training data. If real datasets underrepresent certain populations or encode historical inequities in care delivery, synthetic data may reproduce or even amplify these patterns. This poses a particular challenge for efforts to use synthetic data as a fairness-enhancing tool, as poorly designed generation processes could mask underlying disparities rather than correct them. Rigorous evaluation across demographic and clinical subgroups is therefore essential before synthetic data is used to support clinical AI development (Gonçalves et al., 2020).</p><p>Privacy protection, while often cited as a strength of synthetic data, is not guaranteed by default. Research has shown that some generative models can memorize rare or unique records, leading to potential information leakage if outputs are insufficiently evaluated. El Emam et al. (2023) emphasize the importance of formal disclosure risk assessment and validation frameworks to ensure that synthetic datasets do not enable re-identification attacks. In some cases, integrating techniques such as differential privacy may be necessary to provide mathematically grounded privacy guarantees, particularly when synthetic data is derived from small or sensitive datasets.</p><p><strong>Conclusion</strong></p><p>Ultimately, synthetic data is most powerful not as a complete replacement for real patient records, but as a vital, scalable complement. By embracing hybrid strategies that pair synthetically augmented datasets with real-world clinical validation, we can finally achieve the massive scalability that AI demands while maintaining an ironclad commitment to patient privacy.</p><p>Generative methods are maturing rapidly, and as our evaluation standards become even more rigorous, synthetic data will undoubtedly be the engine driving responsible healthcare innovation. But technology is only half the equation; the real differentiator is the people building it.</p><p>For data scientists, AI engineers, and clinicians looking to make a tangible, life-saving impact, this is the ultimate frontier. We are no longer just theorizing about the future of medicine — we are actively generating the tools to build it. At Philips, we are always looking for the brightest, most innovative minds to help us write this next chapter.</p><p><a href="https://www.careers.philips.com/global/en/philips-tech-careers">Learn more about career opportunities and open roles at Philips here.</a></p><p><em>Philips views and approach are based, in some part, on considering the issues raised by others through their publications as indicated.</em></p><p><strong>References</strong></p><p>Chen, R. J., Lu, M. Y., Chen, T. Y., Williamson, D. F. K., &amp; Mahmood, F. (2023). Synthetic data in machine learning for medicine and healthcare. Nature Biomedical Engineering, 7(6), 493–507. <a href="https://www.nature.com/articles/s41551-021-00751-8">https://www.nature.com/articles/s41551-021-00751-8</a></p><p>El Emam, K., Mosquera, L., &amp; Bass, J. (2023). Evaluating identity disclosure risk in fully synthetic health data: Model development and validation. Journal of Medical Internet Research, 25, e43777. <a href="https://aging.jmir.org/2023/1/e43777">https://aging.jmir.org/2023/1/e43777</a></p><p>Gonçalves, A., Ray, P., Soper, B., Stevens, J., Coyle, L., &amp; Sales, A. P. (2020). Generation and evaluation of synthetic patient data. BMC Medical Research Methodology, 20(1), 108. <a href="https://link.springer.com/article/10.1186/s12874-020-00977-1">https://link.springer.com/article/10.1186/s12874-020-00977-1</a></p><p>Gonzales, A., Guruswamy, G., Smith, S.R. (2023). Synthetic data in health care: A narrative review. PLOS 2(1): e0000082. <a href="https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0000082">https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0000082</a></p><p>Hartjes, M. (2025). Synthetic medical imaging: accelerating AI in healthcare while protecting privacy. Philips. <a href="https://www.philips.com/a-w/about/news/archive/standard/news/articles/2025/synthetic-medical-imaging-accelerating-ai-in-healthcare-while-protecting-privacy.html">https://www.philips.com/a-w/about/news/archive/standard/news/articles/2025/synthetic-medical-imaging-accelerating-ai-in-healthcare-while-protecting-privacy.html</a></p><p>Lemmens, P., Husen, S.A., Rietbergen, M., van Wissen, A. (2024). Synthetic data for fairer AI systems. Aequitas. <a href="https://www.aequitas-project.eu/news/synthetic-data-for-fairer-ai-systems/">https://www.aequitas-project.eu/news/synthetic-data-for-fairer-ai-systems/</a></p><p>van Velzen, M., van der Willigen, R. F., de Beer, V. J., de Graaf-Waar, H. I., Janssen, E. R. C., van Leeuwen, S., van der Willigen, M. F., van der Willigen, M. J., Renardus, G., El Maaroufi, R., Satimin, S. J., Hartog, L. M., Hulsen, T., van Meeteren, N. L. U., &amp; Scheper, M. C. (2025). Privacy-, linguistic-, and information-preserving synthesis of clinical documentation through generative agents. Frontiers in Artificial Intelligence, 8, 1644084. <a href="https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2025.1644084/full">https://www.frontiersin.org/journals/artificial-intelligence/articles/10.3389/frai.2025.1644084/full</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1a8668648948" width="1" height="1" alt=""><hr><p><a href="https://medium.com/philips-technology-blog/synthetic-data-in-healthcare-1a8668648948">Synthetic Data in Healthcare</a> was originally published in <a href="https://medium.com/philips-technology-blog">Philips Technology Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From Unexpected Problems to Predictable Excellence: Why Six Sigma is an Engineer’s Superpower]]></title>
            <link>https://medium.com/philips-technology-blog/from-unexpected-problems-to-predictable-excellence-why-six-sigma-is-an-engineers-superpower-aa5e3382a764?source=rss-bb923c920c2e------2</link>
            <guid isPermaLink="false">https://medium.com/p/aa5e3382a764</guid>
            <dc:creator><![CDATA[Philips]]></dc:creator>
            <pubDate>Wed, 26 Nov 2025 02:56:00 GMT</pubDate>
            <atom:updated>2025-11-26T02:56:00.223Z</atom:updated>
            <content:encoded><![CDATA[<p><em>By Sonia Carbajo, Six Sigma Transformation Leader at Philips</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b0Pmou1gtx1ynkGFKQXAcw.jpeg" /></figure><p>The true power of a Six Sigma mindset lies in its ability to uncover the root cause of a problem, which is often hiding in the one place no one thinks to look. I learned this lesson firsthand on a project I’ll never forget. We had a critical component in one of our systems with a stubbornly high defect rate. Our team of brilliant, dedicated engineers had tried everything. We redesigned the part, we stress-tested it, and tweaked the materials. Nothing worked. The frustration was mounting. Then, a Six Sigma Black Belt joined the project. He didn’t start by looking at the component drawings. Instead, he started with the data for the entire process, from manufacturing to delivery.</p><p>We discovered that the problem wasn’t our design at all. It was the packaging. A simple, data-driven change to how the part was shipped made the problem disappear from our defect list.</p><p>That was my first “aha!” moment with Six Sigma. It showed me that the most complex engineering problems often have surprising root causes, and only a structured, holistic approach can uncover them. Now, with over 20 years of experience as a Six Sigma Transformation Leader at Philips, I play a pivotal role in driving Design for Six Sigma (DFSS) capabilities within Philips’ Innovation Organizations globally, collaborating closely with executive champions, business leaders, and stakeholders.</p><p>Over the years, I have successfully led Lean and Six Sigma transformations in multiple areas — from innovation sites and market organizations to product lines — always focused on embedding continuous improvement, standardizing processes, and empowering future generations through coaching and mentoring.</p><h3>Learning to See the Whole Picture</h3><p>Formally, Six Sigma is a disciplined, data-driven methodology for eliminating defects and variation in any process. Practitioners develop their skills through a certification system inspired by martial arts. Green Belts learn to lead improvement projects within their teams, while Black Belts take on more complex, cross-functional challenges. Master Black Belts focus on coaching these practitioners and driving the overall quality strategy.</p><p>But beyond the formal definitions, Six Sigma is a mindset for delivering excellence. For an engineer, it’s about learning to see the entire system, not just the component in front of you.</p><p>I recall a pivotal project where we were redesigning a part for a system used continuously in the operating room. Our goal was clear: deliver a more robust and reliable product for the healthcare professionals who depend on it. Naturally, our initial focus was on the component’s design. It wasn’t until a Green Belt joined the team that our perspective shifted. By analyzing the data, we uncovered a subtle interaction with how the component was integrated and used in the overall system. After applying Six Sigma methods to address these factors, the redesigned part exceeded expectations, delivering the reliability our customers needed.</p><p>Another moment that stayed with me was the first time we conducted a full value stream map of our factory. We included not just production, but logistics as well. It brought together stakeholders from different disciplines who had never seen the full picture from end to end. People from planning, production, and warehousing saw how their work directly impacted each other and even our supplier collaborations. What started as a process-mapping exercise became a strategic roadmap. It aligned cross-functional teams around key priorities and sparked a series of improvements across the entire operation.</p><p>Sometimes, this mindset shows up in unexpected places. We were improving one of our Design for Six Sigma (DFSS) training exercises, and three Master Black Belts (myself included) spent an afternoon trystorming and measuring the tiniest details. From the outside, it might have looked like overkill. But it reflected exactly what Six Sigma is about: understanding variation, validating a design, and making sure an experience is repeatable and reliable, whether for a customer using our product or an engineer learning a new skill.</p><h3>How We Build Excellence at Philips</h3><p>My journey reflects a broader cultural commitment at Philips. We are embedding this problem-solving mindset across our entire innovation organization. Our approach is twofold.</p><p>Internally, we build a strong foundation. We have a clear change strategy, standardize our processes with a daily management approach, and continuously improve our training. Our Six Sigma Champions play a crucial role here, building competencies and driving alignment across businesses.</p><p>Externally, this internal focus translates directly to our products and services. We emphasize Design for Six Sigma, which means we proactively build quality and reliability into products from the start. We don’t just fix problems; we design for robustness to ensure issues don’t arise in the first place. This approach means investing in our people. We build competencies among key roles (system engineers, industrialization engineers, and quality engineers) so they can execute their processes effectively and in full compliance. This work helps us reduce scrap costs, minimize warranty claims, and improve the customer experience. Guiding teams and future Six Sigma Green Belts and Black Belts is incredibly rewarding, as I see firsthand how it empowers people to innovate with confidence.</p><h3>Why This Matters More Than Ever</h3><p>This way of working feels more relevant today than ever. The challenges modern engineers face demand a more structured and data-informed approach.</p><p>Customer expectations for flawless products are at an all-time high, and innovation cycles are accelerating. The temptation is to move faster, but in healthcare, speed cannot come at the expense of quality and safety. Six Sigma provides the framework to innovate quickly <em>and</em> reliably. It helps teams identify and focus on the critical factors that matter most to the customer, ensuring that speed leads to value without defects.</p><p>For example, consider a team rushing a software update for a medical imaging device. Without a structured process, they might focus only on testing the new feature. But a Six Sigma approach forces them also to consider quality-critical characteristics. Understanding quality-critical parameters is key to making decisions based on data and testing the correct thing at the correct level. It is not about testing more; it is about understanding all the interactions of the system and being able to predict how the system will behave after a change.</p><p>What also makes Six Sigma important now is that we are collecting more data than ever before. Without a structured approach, that data is just noise. A key part of the Six Sigma discipline is turning that data into meaningful insights. This helps us understand process performance, identify sources of variation, and make informed decisions that improve the quality of our products and services.</p><p>Beyond the metrics and processes, this mindset makes the engineering process more rewarding. It shifts the focus from constant firefighting to creative, proactive problem-solving. It gives teams a common language and a proven method to tackle ambiguity, which reduces uncertainty and allows them to focus on what they do best: building incredible products that people trust.</p><p>Consider the difference in how two teams might handle a product failing a late-stage test. One team rushes to implement a fix based on their best guess. They might patch the immediate issue, but they haven’t found the root cause, and they risk shipping a product with a hidden flaw. Another team pauses to collect data, analyze the conditions of the failure, and pinpoint the true source of variation. Instead of just fixing the symptom, they engineer a robust solution and improve the process to prevent it from happening again. That is the difference between hoping for quality and building it with confidence.</p><h3>A Culture of Quality</h3><p>Six Sigma is not a set of tools we occasionally use; it’s a shared mindset at Philips that drives our commitment to quality and innovation. It’s about curiosity, rigor, and a relentless desire to understand the root cause of a problem and ensure that our products are built in a robust way. It’s this culture of continuous improvement and deep problem-solving that makes me proud to be an engineer here. If you are a builder who shares that passion for excellence and wants to help innovate the future of healthcare, you’ll find a home here. <a href="https://www.careers.philips.com/global/en/philips-tech-careers">Learn more about opportunities and roles at Philips here.</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=aa5e3382a764" width="1" height="1" alt=""><hr><p><a href="https://medium.com/philips-technology-blog/from-unexpected-problems-to-predictable-excellence-why-six-sigma-is-an-engineers-superpower-aa5e3382a764">From Unexpected Problems to Predictable Excellence: Why Six Sigma is an Engineer’s Superpower</a> was originally published in <a href="https://medium.com/philips-technology-blog">Philips Technology Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From whiteboard to digital twin: taming complexity in MedTech with Model-Based Systems Engineering]]></title>
            <link>https://philips.medium.com/from-whiteboard-to-digital-twin-taming-complexity-in-medtech-with-model-based-systems-engineering-1e478420b20f?source=rss-bb923c920c2e------2</link>
            <guid isPermaLink="false">https://medium.com/p/1e478420b20f</guid>
            <dc:creator><![CDATA[Philips]]></dc:creator>
            <pubDate>Wed, 22 Oct 2025 23:44:46 GMT</pubDate>
            <atom:updated>2025-10-22T23:44:46.417Z</atom:updated>
            <content:encoded><![CDATA[<p>By Valentina Lavezzo, Director Model-Based Systems Engineering — I&amp;D.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/875/0*tZhee1vQBKbMAHE5.jpeg" /></figure><p>A ​​decade ago, my team ​​and I could map the entire architecture of a new medical system on a single, large whiteboard. We could stand around it, trace connections with a marker, and hold most of the critical interdependencies in our heads. Today, the architecture for a system like our ​​<a href="https://www.philips.com.sg/healthcare/brand/azurion-image-guided-therapy-system">Azurion ImageGuided Therapy </a>or ​​<a href="https://www.philips.com.sg/healthcare/diagnostic-imaging/magnetic-resonance-imaging">Magnetic Resonance Imaging </a>is a network of hardware and software so interconnected that no single mind could fully contain it.</p><p>​​​This growth in complexity is the defining challenge of modern product development, driven by escalating customer expectations, faster time-to-market, and stringent regulatory demands. The reason increased complexity is becoming a challenge is that medical systems are traditionally captured in disconnected tools and documents, leaving the burden of mapping relationships between the documents on human memory. In MedTech, where patient safety is paramount, we cannot afford our development methods to lag behind our product ambitions. We need a more rigorous and connected approach to engineering. This is why we use Model-Based Systems Engineering (MBSE) as the foundational backbone to build a comprehensive digital thread for our products.</p><h3>What is MBSE? More than just a diagram</h3><p>​​​​​Think of developing a complex solution in which software and hardware need to be seamlessly connected. You might have components specified in one tool, an architecture definition in another and requirements at different levels spread across dozens of separate documents. When a team decides to change a component ​during design or for a product already released in the market ​(like a sensor), they must manually hunt down every dependent sub-system or software module that needs modification to prevent a cascade of failures or endangering patient safety. The process is decentralized and highly prone to error.</p><p>MBSE transforms this by creating a single source of truth: a unified digital model of the entire system. In this model, the component is an object with defined interfaces and explicit relationships to every sub-system or software that utilizes it. A proposed change to this component (for example, due to the obsolescence of a part) instantly and automatically flags every point of impact across the system. ​​You can run a​ simulation of the system’s behavior to identify and address possible risks before it is released.​</p><p>This ability to perform virtual integration and analysis directly within the model is a powerful expression of the “shift left” principle. It allows us to identify and resolve potential design and integration flaws at the architectural stage, long before they become expensive and time-consuming problems to fix in the hardware or code.</p><h3>Taming complexity</h3><p>Many engineering organizations have experienced the “hero architect”: a brilliant individual who holds the entire system architecture in their head. These experts are invaluable, but for a modern medical system with over 40 interacting subsystems offered in hundreds of configurations, relying on them creates a significant organizational risk. Knowledge becomes siloed. When that person is unavailable, progress can stall, and design decisions can lack the context needed for effective evolution.</p><p>At Philips, we see MBSE as ​one of ​the strategic solution​s​ to this challenge. It is not about replacing experts; it is about capturing their deep knowledge and democratizing it. The model becomes a shared brain for the team, documenting not just what the system does, but what lies behind the approach to its design. It enables clear communication between teams that are globally located and/or focus on different domains, as the model becomes the single source of truth.</p><p>This approach fundamentally improves the onboarding experience. Instead of spending months trying to build a mental model from scattered documents, new engineers can explore the digital model as an interactive, authoritative guide. They can see the impact of their changes and contribute with confidence far sooner.</p><p>​​ ​This same digital representation creates the resilient, traceable thread from high-level observations down to a specific software requirement, to the component that implements it, and all the way to its verification test and the documentation for our regulatory submissions. Moreover, the labor-intensive process of writing (and maintaining!) the system documentation can now be replaced by a process generating documentation from the system model that is always up-to-date, thereby enabling systems engineers to do what they love: engineering high-quality systems.</p><h3>Putting theory into practice: MBSE at Philips</h3><p>​​ ​Adopting MBSE requires more than just new software; it demands a cultural shift. At Philips, we deploy this through a combination of a central Excellence team, which develops best practices and tools, and MBSE specialists embedded directly within our product development teams. This hybrid approach ensures both consistency across the organization and deep expertise where it is needed most.</p><p>The results are tangible. Imagine, for example, that a late-stage change request comes in from a regulatory body for a patient monitoring system. In a document-based environment, the change impact analysis can take weeks of manual review across dozens of specification documents. When the system is modeled, our team can run digital impact analyses, pinpoint every affected hardware and software module, and deliver a validated implementation plan to management within a few days. Modern systems engineering enables this degree of speed and confidence.</p><h3>The next frontier: powering engineering AI with MBSE</h3><p>The true value of our investment in MBSE extends into the future. Effective AI needs high-quality, structured, and context-rich data to reason effectively. Generic AI cannot understand the intricate dependencies of a medical device. Our system models are based on a curated data set. They provide the context, relationships, and constraints that will allow AI to perform sophisticated engineering analyses.</p><p>We are building toward a future where an engineer can interact with an AI co-pilot built on this foundation. They will be able to ask counterfactual and complex questions, like: ​​”If we replace this processor with a more power-efficient model, what are the cascading impacts on system thermal performance, data throughput, software driver compatibility, and our existing FDA submission evidence?” MBSE is creating the foundational language for us to have that conversation, augmenting engineering craftsmanship with powerful inquiry. AI co-pilots that leverage system models can help us bring the impacts and alternatives to the surface much faster.</p><h3>Building the future of health technology</h3><p>Managing complexity is the new baseline for innovation. By embracing MBSE, we are not just creating better diagrams; we are building a more resilient, agile, and intelligent engineering ​​organization with a digital backbone that supports skilled product developers and amplifies their expertise. This digital foundation allows us to manage today’s complex systems while simultaneously preparing for a future where AI augments our ability to innovate. All this work serves a single purpose: to accelerate the delivery of safe, effective, and life-changing health technology to the patients and clinicians who depend on us.</p><p>If you are an engineer, architect, or developer passionate about solving complex, meaningful problems, we invite you to explore a career at Philips and help us transform the future of healthcare.</p><p>Special thanks to Mike van Diest and Sonja Aarts for reviewing this article before its publication.</p><p>We invite you to explore a career at Philips and help us transform the future of healthcare: <a href="https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.careers.philips.com%2Fglobal%2Fen%2Fphilips-tech-careers&amp;data=05%7C02%7CMariaEduarda.Martins%40philips.com%7Cc367f1ccd8aa479a46f008de106f2ed7%7C1a407a2d76754d178692b3ac285306e4%7C0%7C0%7C638966266869073626%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&amp;sdata=A9UPRoQHowEcMl%2FcwKBuPj249xseHanVYByXfPbtHFc%3D&amp;reserved=0">https://www.careers.philips.com/global/en/philips-tech-careers</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1e478420b20f" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[From whiteboard to digital twin: taming complexity in MedTech with Model-Based Systems Engineering]]></title>
            <link>https://medium.com/philips-technology-blog/from-whiteboard-to-digital-twin-taming-complexity-in-medtech-with-model-based-systems-engineering-c734ca6e5bfb?source=rss-bb923c920c2e------2</link>
            <guid isPermaLink="false">https://medium.com/p/c734ca6e5bfb</guid>
            <dc:creator><![CDATA[Philips]]></dc:creator>
            <pubDate>Tue, 21 Oct 2025 02:16:37 GMT</pubDate>
            <atom:updated>2025-10-22T02:54:02.255Z</atom:updated>
            <content:encoded><![CDATA[<p>By Valentina Lavezzo, Director Model-Based Systems Engineering — I&amp;D.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DohFKzrrLNZLZKc6LXfy0w.jpeg" /></figure><p>A ​​decade ago, my team ​​and I could map the entire architecture of a new medical system on a single, large whiteboard. We could stand around it, trace connections with a marker, and hold most of the critical interdependencies in our heads. Today, the architecture for a system like our ​​<a href="https://www.philips.com.sg/healthcare/brand/azurion-image-guided-therapy-system">Azurion ImageGuided Therapy </a>or ​​<a href="https://www.philips.com.sg/healthcare/diagnostic-imaging/magnetic-resonance-imaging">Magnetic Resonance Imaging </a>is a network of hardware and software so interconnected that no single mind could fully contain it.</p><p>​​​This growth in complexity is the defining challenge of modern product development, driven by escalating customer expectations, faster time-to-market, and stringent regulatory demands. The reason increased complexity is becoming a challenge is that medical systems are traditionally captured in disconnected tools and documents, leaving the burden of mapping relationships between the documents on human memory. In MedTech, where patient safety is paramount, we cannot afford our development methods to lag behind our product ambitions. We need a more rigorous and connected approach to engineering. This is why we use Model-Based Systems Engineering (MBSE) as the foundational backbone to build a comprehensive digital thread for our products.</p><h3>What is MBSE? More than just a diagram</h3><p>​​​​​Think of developing a complex solution in which software and hardware need to be seamlessly connected. You might have components specified in one tool, an architecture definition in another and requirements at different levels spread across dozens of separate documents. When a team decides to change a component ​during design or for a product already released in the market ​(like a sensor), they must manually hunt down every dependent sub-system or software module that needs modification to prevent a cascade of failures or endangering patient safety. The process is decentralized and highly prone to error.</p><p>MBSE transforms this by creating a single source of truth: a unified digital model of the entire system. In this model, the component is an object with defined interfaces and explicit relationships to every sub-system or software that utilizes it. A proposed change to this component (for example, due to the obsolescence of a part) instantly and automatically flags every point of impact across the system. ​​You can run a​ simulation of the system’s behavior to identify and address possible risks before it is released.​</p><p>This ability to perform virtual integration and analysis directly within the model is a powerful expression of the “shift left” principle. It allows us to identify and resolve potential design and integration flaws at the architectural stage, long before they become expensive and time-consuming problems to fix in the hardware or code.</p><h3>Taming complexity</h3><p>Many engineering organizations have experienced the “hero architect”: a brilliant individual who holds the entire system architecture in their head. These experts are invaluable, but for a modern medical system with over 40 interacting subsystems offered in hundreds of configurations, relying on them creates a significant organizational risk. Knowledge becomes siloed. When that person is unavailable, progress can stall, and design decisions can lack the context needed for effective evolution.</p><p>At Philips, we see MBSE as ​one of ​the strategic solution​s​ to this challenge. It is not about replacing experts; it is about capturing their deep knowledge and democratizing it. The model becomes a shared brain for the team, documenting not just what the system does, but what lies behind the approach to its design. It enables clear communication between teams that are globally located and/or focus on different domains, as the model becomes the single source of truth.</p><p>This approach fundamentally improves the onboarding experience. Instead of spending months trying to build a mental model from scattered documents, new engineers can explore the digital model as an interactive, authoritative guide. They can see the impact of their changes and contribute with confidence far sooner.</p><p>​​ ​This same digital representation creates the resilient, traceable thread from high-level observations down to a specific software requirement, to the component that implements it, and all the way to its verification test and the documentation for our regulatory submissions. Moreover, the labor-intensive process of writing (and maintaining!) the system documentation can now be replaced by a process generating documentation from the system model that is always up-to-date, thereby enabling systems engineers to do what they love: engineering high-quality systems.</p><h3>Putting theory into practice: MBSE at Philips</h3><p>​​ ​Adopting MBSE requires more than just new software; it demands a cultural shift. At Philips, we deploy this through a combination of a central Excellence team, which develops best practices and tools, and MBSE specialists embedded directly within our product development teams. This hybrid approach ensures both consistency across the organization and deep expertise where it is needed most.</p><p>The results are tangible. Imagine, for example, that a late-stage change request comes in from a regulatory body for a patient monitoring system. In a document-based environment, the change impact analysis can take weeks of manual review across dozens of specification documents. When the system is modeled, our team can run digital impact analyses, pinpoint every affected hardware and software module, and deliver a validated implementation plan to management within a few days. Modern systems engineering enables this degree of speed and confidence.</p><h3>The next frontier: powering engineering AI with MBSE</h3><p>The true value of our investment in MBSE extends into the future. Effective AI needs high-quality, structured, and context-rich data to reason effectively. Generic AI cannot understand the intricate dependencies of a medical device. Our system models are based on a curated data set. They provide the context, relationships, and constraints that will allow AI to perform sophisticated engineering analyses.</p><p>We are building toward a future where an engineer can interact with an AI co-pilot built on this foundation. They will be able to ask counterfactual and complex questions, like: ​​”If we replace this processor with a more power-efficient model, what are the cascading impacts on system thermal performance, data throughput, software driver compatibility, and our existing FDA submission evidence?” MBSE is creating the foundational language for us to have that conversation, augmenting engineering craftsmanship with powerful inquiry. AI co-pilots that leverage system models can help us bring the impacts and alternatives to the surface much faster.</p><h3>Building the future of health technology</h3><p>Managing complexity is the new baseline for innovation. By embracing MBSE, we are not just creating better diagrams; we are building a more resilient, agile, and intelligent engineering ​​organization with a digital backbone that supports skilled product developers and amplifies their expertise. This digital foundation allows us to manage today’s complex systems while simultaneously preparing for a future where AI augments our ability to innovate. All this work serves a single purpose: to accelerate the delivery of safe, effective, and life-changing health technology to the patients and clinicians who depend on us.</p><p>If you are an engineer, architect, or developer passionate about solving complex, meaningful problems, we invite you to explore a career at Philips and help us transform the future of healthcare.</p><p>Special thanks to Mike van Diest and Sonja Aarts for reviewing this article before its publication.</p><p>We invite you to explore a career at Philips and help us transform the future of healthcare: <a href="https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.careers.philips.com%2Fglobal%2Fen%2Fphilips-tech-careers&amp;data=05%7C02%7CMariaEduarda.Martins%40philips.com%7Cc367f1ccd8aa479a46f008de106f2ed7%7C1a407a2d76754d178692b3ac285306e4%7C0%7C0%7C638966266869073626%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&amp;sdata=A9UPRoQHowEcMl%2FcwKBuPj249xseHanVYByXfPbtHFc%3D&amp;reserved=0">https://www.careers.philips.com/global/en/philips-tech-careers</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c734ca6e5bfb" width="1" height="1" alt=""><hr><p><a href="https://medium.com/philips-technology-blog/from-whiteboard-to-digital-twin-taming-complexity-in-medtech-with-model-based-systems-engineering-c734ca6e5bfb">From whiteboard to digital twin: taming complexity in MedTech with Model-Based Systems Engineering</a> was originally published in <a href="https://medium.com/philips-technology-blog">Philips Technology Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The AI Revolution in Healthcare]]></title>
            <link>https://medium.com/philips-technology-blog/the-ai-revolution-in-healthcare-2fe341229ac2?source=rss-bb923c920c2e------2</link>
            <guid isPermaLink="false">https://medium.com/p/2fe341229ac2</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[big-data]]></category>
            <category><![CDATA[healtchare]]></category>
            <category><![CDATA[ai]]></category>
            <dc:creator><![CDATA[Philips]]></dc:creator>
            <pubDate>Mon, 06 Jan 2025 15:02:36 GMT</pubDate>
            <atom:updated>2025-01-06T15:02:36.600Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Authors: Tim Hulsen, </strong>Senior Data &amp; AI Scientist at Philips &amp; Professor AI &amp; Data Supported Healthcare, Rotterdam University of Applied Sciences, <strong>Francesca Manni, </strong>Clinical Scientist at Philips and <strong>Katarzyna Markiewicz</strong>, Director, AI Strategic Intelligence Lead at Philips</p><p>Artificial intelligence and big data analytics are ushering in a transformative era in healthcare, fundamentally changing our approach to public health. These technologies enable healthcare providers to move from reactive to preventive care, analyze vast amounts of medical data in real time, and deliver more personalized treatment options to patients worldwide. Healthcare providers are becoming increasingly aware of the benefits and opportunities of applying AI in the patient care cycle [1].</p><p>According to the Philips 2024 Future Health Index [2], a comprehensive study of 3,000 healthcare leaders across 14 countries, healthcare leaders have already implemented AI for clinical decision support across different domains, e.g., radiology and in-hospital patient monitoring. Furthermore, 92% of leaders consider automation crucial for addressing staff shortages and freeing up more time for high-skill work and enabling better patient care. Looking ahead, 41% of healthcare leaders prioritize AI implementation in remote patient monitoring over the next three years.</p><p>However, despite the AI hype observed for several years and the fact that, like many digital health applications, AI thrived during the pandemic, we can safely say that we are still at relatively early stages of AI maturity [3]. Growing healthcare challenges combined with opportunities that AI technologies bring, creates a huge opportunity for talented machine learning and AI professionals who want to put their skills to good use to help improve the lives of people across the world.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*b0Pmou1gtx1ynkGFKQXAcw.jpeg" /><figcaption>Source: Philips</figcaption></figure><h3>Emerging Trends Shaping Healthcare AI</h3><p>The healthcare AI landscape is rapidly evolving across two major fronts. Traditional ML continues to advance diagnostic accuracy, optimize clinical workflows, and enhance medical imaging analysis. Meanwhile, generative AI is opening new frontiers in drug discovery, medical research synthesis, and personalized treatment planning. These developments are creating unprecedented opportunities for AI practitioners to drive healthcare innovation [4].</p><h3>Philips’ Vision for AI-Powered Healthcare</h3><p>Philips champions a human-centered approach to AI development, creating solutions that enhance rather than replace healthcare professionals, enabling them to focus on patient care. Our AI initiatives in medical imaging, patient monitoring, and healthcare informatics are guided by rigorous ethical principles and a singular focus: improving patient outcomes worldwide. AI and big data are some of the tools that are helping us achieve our vision to provide care to billions of people.</p><h3>Philips’ approach</h3><h3>Purpose-Driven Innovation</h3><p>AI represents more than a path to profit — we believe it’s a powerful tool for improving human lives. Our innovations help healthcare providers deliver better care to more patients worldwide.</p><h3>Real-World Impact Through Global Partnerships</h3><p>Our vast partnerships with healthcare providers and hospitals worldwide provide invaluable real-world feedback and insights. AI practitioners at Philips work on actual healthcare challenges, collaborating directly with medical professionals who use our solutions daily.</p><h3>Diverse, High-Quality Data Access</h3><p>Our global presence provides access to diverse and high-quality medical data sets, critical for developing more inclusive and unbiased AI systems [5]. This unique advantage enables our teams to create solutions that work effectively across different demographics and healthcare settings.</p><h3>Responsible AI</h3><p>While we are pushing the boundaries of what is possible at the intersection of AI and healthcare, we don’t lose sight of ethical principles such as transparency, reliability, and fairness. Responsible AI (RAI) provides a framework to ensure these principles are preserved in AI initiatives [6]. By embracing the (current and future) technical possibilities of AI, and at the same time making sure that AI is explainable and responsible, we help hospitals withstand future challenges [7].</p><h3>World-Class Expertise and Collaboration</h3><p>Philips brings together leading experts in healthcare and technology. Our collaborative environment ensures that AI specialists can focus on their strengths while learning from professionals across various disciplines, from clinical experts to software architects.</p><h3>Philips AI projects</h3><p>Our AI solutions are carefully designed to give back time to healthcare professionals and enhance their decision-making. Whether by enabling fast, high-quality imaging, optimizing clinical workflows, or providing timely clinical insights, we use AI to address providers’ most pressing needs and challenges. By integrating AI into clinical and operational workflows, we enable more efficient, informed, and personalized care. We focus on enhancing human expertise and empowering caregivers to focus on what matters most: patient care. Our AI solutions are centered around four value drivers: 1) streamlining workflows and operations; 2) providing clinical insights; 3) expanding access to patient care; and 4) supporting personalized self-care.</p><p>Here are some examples of Philips AI solutions related to these value drivers:</p><ol><li>Streamlining workflows and operations</li></ol><p>Philips AI solutions automate administrative and repetitive tasks, giving healthcare professionals more time to focus on patient care. For example, <strong>MR SmartSpeed</strong> [8] is an AI-based imaging technology that can increase imaging speed by up to a factor of 3 while providing up to 65% greater resolution* to deliver outstanding image quality. It is applicable to 97% of clinical protocols** to address the needs of a broad range of patients in various conditions.</p><p>*Compared to SENSE imaging.</p><p>**On average, measured across a sample of sites from Philips MR Installed Base.</p><p>2. Providing clinical insights</p><p>Philips AI solutions empower clinicians with actionable insights at the point of care, supporting informed decision-making in critical moments. Our <strong>CT Lung Nodule CAD</strong> [9] offers an automated process that identifies and marks regions of interest based on image features associated with lung nodules. It is intended for use as a second reader after an initial interpretation of the diagnostic image has been performed [10].</p><p>3. Expanding access to patient care</p><p>Philips AI solutions help healthcare professionals reach more patients, including those in underserved communities, by extending high-quality care from the hospital into the home and other care settings. One example is <strong>Cardiologs</strong> [11], an arrhythmia diagnostic software that is cloud-based, vendor-neutral and powered by AI* to streamline ECG analysis. Its clinically validated algorithm helps detect more than 20 types of arrhythmia events and parameters to expand the breadth of the physician’s clinical interpretation.</p><p>*The Cardiologs Platform interpretation results are not intended to be the sole means of diagnosis. It is offered to physicians and clinicians on an advisory basis only in conjunction with the physician’s knowledge of ECG patterns, patient background, clinical history, symptoms, and other diagnostic information.</p><p>4. Supporting personalized self-care</p><p>Philips AI solutions help people take care of their health and well-being at home by offering personalized guidance tailored to their needs and preferences. <strong>Philips Sonicare DiamondClean Prestige 9900</strong> [12] is an electric power toothbrush that uses sensors to detect the cleaning motions people use and the coverage they achieve. It pairs with a mobile app that allows people to better understand their brushing technique and take an active role in improving their oral health.</p><p><strong><em>Curious about working in tech at Philips? </em></strong><a href="https://www.careers.philips.com/global/en/philips-tech-careers"><strong><em>Find out more here</em></strong></a></p><p>References:</p><p>1. Luordo, D. et al.: Artificial Intelligence and Healthcare: A Journey through History, Present Innovations, and Future Possibilities. Life 14 (5): 557 (2024). <a href="https://www.mdpi.com/2075-1729/14/5/557">https://www.mdpi.com/2075-1729/14/5/557</a></p><p>2. Philips 2024 Future Health Index. <a href="https://www.philips.com/a-w/about/news/future-health-index/reports/2024/better-care-for-more-people">https://www.philips.com/a-w/about/news/future-health-index/reports/2024/better-care-for-more-people</a></p><p>3. Olaye, I.M., Seixas, A.A.: The Gap Between AI and Bedside: Participatory Workshop on the Barriers to the Integration, Translation, and Adoption of Digital Health Care and AI Startup Technology Into Clinical Practice. J. Med. Internet Res. 25, e32962 (2023). <a href="https://www.jmir.org/2023/1/e32962/">https://www.jmir.org/2023/1/e32962/</a></p><p>4. Hulsen, T.: Artificial Intelligence in Healthcare: Current State and Future Perspectives<strong>. </strong>MDPI (2024). <a href="https://www.mdpi.com/books/reprint/9229-artificial-intelligence-in-healthcare-current-state-and-future-perspectives">https://www.mdpi.com/books/reprint/9229-artificial-intelligence-in-healthcare-current-state-and-future-perspectives</a></p><p>5. Hulsen, T.: Big and rich data as fuel for the AI engine. Medium (2023). <a href="https://medium.com/philips-technology-blog/big-and-rich-data-as-fuel-for-the-ai-engine-a7c10755b003">https://medium.com/philips-technology-blog/big-and-rich-data-as-fuel-for-the-ai-engine-a7c10755b003</a></p><p>6. Trocin, C., Mikalef, P., Papamitsiou, Z., Conboy, K.: Responsible AI for digital health: A synthesis and a research agenda. Inf. Syst. Front. 25, 2139–2157 (2023). <a href="https://doi.org/10.1007/s10796-021-10146-4">https://doi.org/10.1007/s10796-021-10146-4</a></p><p>7. Kuran, D., van Wissen, A., van Hartskamp, M., Janssen, G.: Navigating the Future: Philips and the Evolving AI Regulation Landscape. <a href="https://medium.com/philips-technology-blog/navigating-the-future-philips-and-the-evolving-ai-regulation-landscape-5ff05e70bd3d">https://medium.com/philips-technology-blog/navigating-the-future-philips-and-the-evolving-ai-regulation-landscape-5ff05e70bd3d</a></p><p>8. Philips MR SmartSpeed. <a href="https://www.usa.philips.com/healthcare/resources/landing/smartspeed">https://www.usa.philips.com/healthcare/resources/landing/smartspeed</a></p><p>9. Philips CT Lung Nodule analysis ClearRead CAD.</p><p><a href="https://www.philips.co.uk/healthcare/product/NICA555/ct-lung-nodule-analysisclearreadcad">https://www.philips.co.uk/healthcare/product/NICA555/ct-lung-nodule-analysisclearreadcad</a></p><p>10. Philips Advanced Visualization Workspace 15 — Clinical Datasheet. <a href="https://www.documents.philips.com/assets/20230725/835aafd3d76e420d987bb04a01540d97.pdf">https://www.documents.philips.com/assets/20230725/835aafd3d76e420d987bb04a01540d97.pdf</a></p><p>11. Philips Cardiologs <a href="https://cardiologs.com/">https://cardiologs.com/</a></p><p>12. Philips AI-enabled Solutions. <a href="https://www.philips.com/a-w/about/artificial-intelligence/ai-enabled-solutions.html#triggername=less2_isc2">https://www.philips.com/a-w/about/artificial-intelligence/ai-enabled-solutions.html#triggername=less2_isc2</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2fe341229ac2" width="1" height="1" alt=""><hr><p><a href="https://medium.com/philips-technology-blog/the-ai-revolution-in-healthcare-2fe341229ac2">The AI Revolution in Healthcare</a> was originally published in <a href="https://medium.com/philips-technology-blog">Philips Technology Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[WebSummit 2024: A wake-up call for healthcare innovation]]></title>
            <link>https://medium.com/philips-technology-blog/websummit-2024-a-wake-up-call-for-healthcare-innovation-65c4ca865d07?source=rss-bb923c920c2e------2</link>
            <guid isPermaLink="false">https://medium.com/p/65c4ca865d07</guid>
            <category><![CDATA[ar]]></category>
            <category><![CDATA[healthtech]]></category>
            <category><![CDATA[cloud]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[healthcare]]></category>
            <dc:creator><![CDATA[Philips]]></dc:creator>
            <pubDate>Thu, 21 Nov 2024 13:50:08 GMT</pubDate>
            <atom:updated>2024-11-26T08:33:23.849Z</atom:updated>
            <content:encoded><![CDATA[<p>Author: Dr. Atul Gupta, Chief Medical Officer of Diagnosis &amp; Treatment at Philips</p><p>Taking the stage at WebSummit 2024 in Lisbon, I was ready to delve into a critical conversation about the pressing challenges facing healthcare today. The world’s healthcare systems are buckling under immense strain. Rising chronic diseases, aging populations, and critical shortages of healthcare workers have brought us to a tipping point. At an event known for its forward-thinking solutions, my focus was to explore how technology can meet these challenges head-on — not just in theory, but in practical and measurable ways.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*hXbT6VxBnLziARO85hTDuw.jpeg" /><figcaption>Dr. Atul Gupta at WebSummit 2024</figcaption></figure><p>For those who aren’t familiar, WebSummit is one of the largest technology gatherings globally, a melting pot of tech people drawing over 70,000 attendees from diverse industries. It’s where innovation meets urgency, a perfect setting to discuss how the healthcare sector can navigate its way out of what often feels like a crisis without an end.</p><p><strong>A Sobering Picture of today’s healthcare</strong></p><p>Global healthcare systems are in trouble. Nearly half of healthcare professionals report burnout, and by 2030, we face an estimated shortfall of 10 million healthcare workers worldwide​​ [1]. In the UK, the National Health Service (NHS) has over 100,000 vacancies, and waiting lists for care are growing longer each year​. And in the US, physician and nurse burnout is at an all-time high — over 50% in some surveys! These aren’t just statistics — they represent delayed diagnoses, missed treatments, and lives lost.</p><p>For years, we’ve tried to “recruit our way out” of this crisis. It’s clear now that this strategy alone will not suffice. What we need is nothing short of a transformation in how we deliver care — one that leverages advanced technologies like artificial intelligence, robotics, and augmented reality to bridge the gaps.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DW69exZHaey2p0b04qTVJg.jpeg" /><figcaption>Dr. Atul Gupta at WebSummit 2024</figcaption></figure><p><strong>The role of technology: today and tomorrow</strong></p><p>At Philips, we’re actively developing and deploying solutions that can make a tangible difference today. For example, in stroke care, time is brain. Delays in diagnosis or treatment lead to devastating consequences, with 2 million neurons lost every minute a stroke goes untreated​​. AI-driven imaging tools, cloud-based telepresence solutions, and AR guidance systems are not just promising ideas — they are actively saving lives by streamlining every step of the stroke care pathway.</p><p>Looking ahead, the future of healthcare could be profoundly reshaped by intelligent systems. Picture an operating suite where AI integrates seamlessly with AR and robotics, enabling physicians to treat complex conditions with precision and speed that would have been unimaginable a decade ago​​. These technologies are not just about convenience; they have the potential to drastically reduce mortality rates, improve outcomes, and make care accessible even in the most underserved areas.</p><p><strong>The urgency of collaboration</strong></p><p>However, technology alone isn’t the answer. The real key to success lies in collaboration. This isn’t a problem for healthcare professionals alone to solve. Clinicians, technologists, policymakers, and even patients, we all need to work together. At the summit, I was particularly struck by the perspectives of young innovators — the millennials and Gen Z’s — who are challenging our traditional ways of thinking. Their ideas are crucial as we move toward a future where technology will be an integral part of healthcare delivery.</p><p>One sobering statistic stayed with me throughout the event: half of the global population still lacks access to basic healthcare​. It’s a reminder that even as we innovate, we must ensure that these advancements are scalable and equitable, reaching those who need them most.</p><p><strong>Why this matters to you</strong></p><p>For those of you early in your careers or considering where to make an impact, I urge you to look closely at healthcare. The challenges are daunting, but so are the opportunities to drive meaningful change. Whether it’s building algorithms to diagnose diseases earlier, designing robotic systems to extend the capabilities of healthcare teams, or rethinking patient workflows, the work you do could touch millions of lives.</p><p>At Philips, we are committed to making healthcare more inclusive, connected, and sustainable. But we can’t do it alone. If you share a vision of transforming care delivery and addressing the inequities in healthcare access, we need your expertise.</p><p>As I said to the audience at Web Summit: <strong>Innovation is not optional — it’s essential. The stakes are too high, and the time to act is now. </strong>Let’s work together to deliver solutions that not only meet today’s challenges but define the future of healthcare.</p><p><strong><em>Curious about working in tech at Philips? </em></strong><a href="https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.careers.philips.com%2Fglobal%2Fen%2Fphilips-tech-careers&amp;data=05%7C02%7Cjoost.maltha%40philips.com%7C46e78d4ab2c54807de7908dd07ac470e%7C1a407a2d76754d178692b3ac285306e4%7C0%7C0%7C638675158686417667%7CUnknown%7CTWFpbGZsb3d8eyJFbXB0eU1hcGkiOnRydWUsIlYiOiIwLjAuMDAwMCIsIlAiOiJXaW4zMiIsIkFOIjoiTWFpbCIsIldUIjoyfQ%3D%3D%7C0%7C%7C%7C&amp;sdata=6Z3wEOgcioAiRYCiPB5QcRyPmqye%2BM7M4n2akSshfJo%3D&amp;reserved=0"><strong><em>Find out more here</em></strong></a></p><p>[1] <a href="https://www.who.int/health-topics/health-workforce?utm_source=chatgpt.com">World Health Organization</a></p><p>[2] <a href="https://www.bma.org.uk/advice-and-support/nhs-delivery-and-workforce/workforce/nhs-medical-staffing-data-analysis?utm_source=chatgpt.com">British Medical Association</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=65c4ca865d07" width="1" height="1" alt=""><hr><p><a href="https://medium.com/philips-technology-blog/websummit-2024-a-wake-up-call-for-healthcare-innovation-65c4ca865d07">WebSummit 2024: A wake-up call for healthcare innovation</a> was originally published in <a href="https://medium.com/philips-technology-blog">Philips Technology Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The Craftsperson’s Toolkit: Professionalism, Pragmatism, and Self-Improvement]]></title>
            <link>https://medium.com/philips-technology-blog/the-craftspersons-toolkit-professionalism-pragmatism-and-self-improvement-e3f5ea64a6ac?source=rss-bb923c920c2e------2</link>
            <guid isPermaLink="false">https://medium.com/p/e3f5ea64a6ac</guid>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[software]]></category>
            <dc:creator><![CDATA[Philips]]></dc:creator>
            <pubDate>Wed, 23 Oct 2024 06:31:47 GMT</pubDate>
            <atom:updated>2024-10-23T06:31:47.469Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Authors: Fernando Vieira and Rodolfo Hansen, Software Compentency Leads (Developer advocates/evangelists)</strong></p><p>Despite the ubiquity of software in modern life, the public often overlooks its complexity. On the one hand, people still view the software industry through the lens of traditional engineering disciplines like mechanical or electrical engineering, where outcomes are more predictable, quietly ignoring how software uniquely explodes in its dynamic complexity. At the same time, it is common to find in media statements like “you can become a millionaire overnight with a killer app after taking this 6-month course”. The combination of both these views inspires young people to pursue careers in software engineering motivated by false assumptions: software engineering is not especially hard, you can master it in 6 months, and you will make your fortunes by coming onboard. This turns into a unique challenge for the industry at large.</p><p>Robert C. Martin (a.k.a. Uncle Bob) has accurately predicted the challenge, including the experience gap (nearly half of software engineers have less than five years of experience) and the growing potential for legal liability for individuals due to the consequences of software failures[1]. To address these issues, he has advocated software engineers should pursue excellence and work with professionalism. Uncle Bob is right. Even with advancements in tools and processes, the quality of software remains heavily reliant on the craft of engineers. And developing craftsmanship requires more than just enthusiasm; it requires a combination of continuous learning, responsibility and commitment to excellence throughout a professional lifetime.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C8JgopwOMM4Jn60YaNtXzA.jpeg" /><figcaption>Source: Philips</figcaption></figure><p>Another factor underestimated by the media and influencers is the vast amount of existing software. Currently, it is unlikely a software engineer will be hired to deal only with new projects. We must acknowledge that today, more than ever, being a software developer means joining a company with an existing codebase. Companies hire developers to add features to their existing products while making improvements in performance, security, reliability, etc. Managers expect newcomers to learn the technology stack previously adopted; one which is often outdated. This means software teams always need to face the problem of technical debt, which is a pervasive issue in the software industry, even in startups.</p><p>This is the new normal of our industry. It is a bit far from idealized portrayals often presented but it is not a catastrophe. We should adapt to the real world and find opportunities with these circumstances. Also, while you are affected by consequences of your predecessor’s decisions, developers will likewise be affected by consequences of your decisions. Our decisions need to embrace responsibility and accountability. The current landscape requires decisions balancing innovation with the maintenance and improvement of existing systems.</p><p>The balance can be reached when teams have maturity in their development practices. Technical debt management works well when the team diligently applies the Boy Scout Rule[2], has a defined process, and can reserve part of its capacity to pay back this debt. Static analysis and strong code review can guarantee that new or modified code meets quality standards. A holistic test approach will provide safety for developers refactoring code and improving design. Strong automation will ensure that developers do not spend time on tasks that a machine could handle. Finally, constructive programming and other formal methods can add safety and correctness to the code the team produces.</p><p>Mastering relevant software practices is essential because they give you freedom. Once a team reaches maturity, developers discover they finally have time to experiment and add new technologies or paradigms. Then, your responsibility as a professional is to understand if you are choosing a solution because you would like to try it or because it is the best option for a given problem. If the decision involves more passion than reason, you should analyze the true value the new item will add to the product you are responsible for. Remember that technology is just a means to an end, and that simple solutions can be very elegant and effective.</p><p>Artificial Intelligence is a good example. No one doubts AI will be integrated with both software products and the development environment. However, software engineers should carefully consider whether AI is the optimal solution when alternatives are available. The decision to incorporate AI into an existing product needs to start with the customer perceiving value on the solution and end up with the company being able to deal with the development and operational complexity AI would add during the lifespan of software.</p><p>The landscape we painted is challenging and imperfect, but it is also full of opportunities. It demands people who are resilient enough to maintain professionalism and commitment with continuous learning even in the face of adversity; it demands people who truly feel passionate about software. By embodying these qualities, they will not only build solid careers; they also can make a positive impact on society through their work in a world increasingly driven by software.</p><p>So how to build a solid career? Do it by everyday endeavors in personal and professional development. Although authors differ in the number and name of categories, the core principles remain consistent: professionalism, pragmatism, and self-improvement. By striving for excellence in these areas, you’ll become a valuable professional to any organization, regardless of the ever-evolving challenges of the software development industry. On the following paragraphs you will find detailed descriptions of each principle, followed by improvement suggestions.</p><p><strong>Professionalism</strong></p><p>A professional is an individual who consistently demonstrates excellence in their field through a combination of technical competence, ethical conduct, and interpersonal skills. Professionals are accountable for their actions, proactive in problem-solving, and committed to continuous learning. They foster a positive and collaborative work environment by exhibiting respect, reliability, a strong work ethic and reciprocity. This means that professionals expect others to uphold the same high standards of behavior that they show.</p><p>A professional developer must:</p><p>● <strong>Master the craft — </strong>Domain development practices, being able to apply them when solving complex problems. The outcome is high quality code that meets all requirements and is well designed, efficient and maintainable.</p><p>● <strong>Demonstrate ethical conduct</strong> — Exhibit integrity, honesty and respect for others. Adhere to laws, code of ethics and company principles, when available. Prioritize society’s best interests.</p><p>● <strong>Communicate effectively</strong> — Read, write and speak clearly and concisely. Learn to effectively translate technical concepts into non-technical language, and vice versa, while communicating with all people involved in a software project.</p><p>How to improve:</p><p>● <strong>Master the Craft</strong> — Learn and practice all development practices through reading books and articles, watching videos and lectures, and exercising with personal and open-source projects.</p><p>● <strong>Demonstrate ethical conduct</strong> — Analyze how your actions impact people. Explore your company’s principles and read codes of conduct like the IEEE one. Understanding how your code impacts customers and society can provide insights into making things better for everyone.</p><p>● <strong>Communicate effectively</strong> — Reading and writing are the cornerstones of effective communication. There is a vast array of online courses focused on developing communication abilities.</p><p><strong>Pragmatism</strong></p><p>Pragmatism[3] is essential for navigating into the current landscape of software development. It does not mean an excuse to avoid pursuing excellence in accepting things worse than they should be. It means understanding the difference between the ideal and real world, to deal with trade-offs, and it means making the best with the available resources.</p><p>In short, pragmatism is about being practical, flexible, and results-oriented. It’s about understanding the realities of the industry, making informed decisions, and delivering the best possible software within the given constraints.</p><p>A pragmatic developer must:</p><p>● <strong>Balance theory and practice </strong>— Master foundational concepts and apply them to real-world scenarios, fostering adaptability to technological changes and building enduring knowledge.</p><p>● <strong>Adhere to YAGNI </strong>— Avoid complex technology or design aiming a possible future. The “you ain’t gonna need it”[4] principle should guide your approach. Refrain from overcomplicating solutions for today’s problems unless there’s a clear business case or a well-defined product roadmap justifying the complexity.</p><p>● <strong>Be adaptable</strong> — Be flexible and able to learn processes and technologies quickly. It is an essential trait the best software companies look for due to the proliferation of programming languages and tools. Be willing to adopt newer technologies when they offer clear advantages, avoiding getting stuck due to personal preference or the time dedicated to learning older ones. Embrace a mindset of continuous learning, recognizing that technology is a tool, and core value lies in problem-solving.</p><p>● <strong>Analyze cost-benefit </strong>— Be able to evaluate the trade-offs and the consequences of a choice. Make the best possible decisions given the limitations.</p><p>How to improve:</p><p>● <strong>Balance theory and practice </strong>— A good exercise you can always do is trying to explain some of your practical decisions with theory. The same applies when you learn new concepts: try to apply them in some practical problem. This way you will make sure that your knowledge is solid.</p><p>● <strong>Adhere to YAGNI </strong>— When choosing technology, try to do score cards with honest criteria. Let the reasoning decide what option best fits with the problem. When designing systems, try to do it incrementally. Test-Driven Development is great because it forces you to gradually adapt your solution to the real cases the software must address.</p><p>● <strong>Be adaptable </strong>— Adopting a positive attitude towards change is the first step. Another crucial aspect is to actively exercise your adaptability skills. You can do this by tackling small coding challenges in programming languages that you haven’t mastered. Try to solve one small problem daily, and you’ll notice a significant improvement in your adaptability.</p><p>● <strong>Analyze cost-benefit </strong>— Start trying to do the analysis for small decisions. You can try techniques like balanced score cards and decision trees. Use the same process to make decisions in your life. For example, to decide what car or mobile phone best fits your needs.</p><p><strong>Self-Improvement</strong></p><p>Self-improvement is a lifelong pursuit of personal growth and development. It involves planning your time wisely, keeping promises and being able to learn and improve your craft. It is about doing your best at work and life, including ways to cope with stress of hard times, balancing both. It is also about finding passion and purpose in a way that your work integrates with your values, being a source of satisfaction and achievements.</p><p>A committed developer must:</p><p>·<strong>Manage time effectively </strong>— Know how to deal with numerous duties and distractions, prioritizing tasks and accomplishing outcomes. Be punctual at meetings; punctuality is essential for maintaining a positive professional image.</p><p>· <strong>Engage in continuous learning </strong>— Expend additional hours to learn. Uncle Bob states we should dedicate an additional 20 hours per week to be up to date with technology and improve our craft. Although the number of hours is debatable, there is no doubt software developers need to make extra effort to have a solid career. If the modern world changes fast, the software industry changes 20 times faster.</p><p>· <strong>Pursue passion and purpose </strong>— Find passion in your work and have a clear sense of purpose, aligning your work with your values. Consider the importance of your work in the daily life of people you know. The comfort of the modern world also relies on software.</p><p>· <strong>Maintain a work-life balance </strong>— Pursue a good life, a life that allows you to spend time with family, friends and things you like while still value the profession.</p><p>· <strong>Deal with stress</strong> <strong>— </strong>Develop effective coping mechanisms for dealing with work-related stress. The question is not if you will experience stress, but when.</p><p>How to improve:</p><p>● <strong>Manage time effectively — </strong>There are many articles, videos and courses about this subject. Here our suggestion is to try some techniques and to embrace the ones that you can naturally incorporate into your daily life. If you work from home, establish a clear boundary between your personal and professional life.</p><p>● <strong>Engage in continuous learning — </strong>Plan your life with continuous learning as an unnegotiable task. You will have to find time to read articles and books, or just to play with some problems using unfamiliar technology. Maintain a notebook to track knowledge gaps and learning goals.</p><p>● <strong>Pursue passion and purpose — </strong>Define what is important to you and the people you love. After, analyze how the products and services you contribute to support what you value. Put yourself in the position of your customer (sometimes it is not even necessary because you are also the customer). Like any customer, you like to get the best possible product or service. This view, alone, gives you the importance of pursuing excellence.</p><p>● <strong>Maintain a work-life balance — </strong>In order to make a good balance, set clear boundaries between work and personal life and communicate it. Having passion and energy in work is great, but spending leisure time at work often leads to burnout. Spare time is fundamental for your health and well-being.</p><p>● <strong>Deal with stress</strong> — It is common for people to engage in mindfulness and relaxation sessions or practicing physical activity. Having hobbies always helps; if you don’t have a hobby, find one. Some developers can combat stress coding for personal or open-source projects; some others prefer to take a break from the tech world. The goal is to alleviate stress and continue your journey toward professional and personal fulfillment.</p><p><strong>A Final Thought</strong></p><p>Remember that circumstances play a big role in our lives. To adapt and evolve is required for each living organism. If you read along and find it overwhelming, try reading again and highlighting any points in the above list that can help you, personally, improve as a craftsperson. Hang on to those and realize that it is indeed possible to grow and flourish in the context of what it means to be a software developer today. There have been plenty of recent advancements that need to make their way into the industry, and you are responsible for making the environment you work in follow and model your own passions.</p><p><strong><em>Curious about working in tech at Philips? </em></strong><a href="https://www.careers.philips.com/global/en/philips-tech-careers"><strong><em>Find out more here</em></strong></a></p><p>References:</p><p>[1] Clean Craftsmanship: Disciplines, Standards, and Ethics (Robert C. Martin)</p><p>[2] Clean Code: A Handbook of Agile Software Craftsmanship (Robert C. Martin)</p><p>[3] The Software Craftsman: Professionalism, Pragmatism, Pride (Sandro Mancuso)</p><p>[4] <a href="https://en.wikipedia.org/wiki/You_aren%27t_gonna_need_it">https://en.wikipedia.org/wiki/You_aren%27t_gonna_need_it</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=e3f5ea64a6ac" width="1" height="1" alt=""><hr><p><a href="https://medium.com/philips-technology-blog/the-craftspersons-toolkit-professionalism-pragmatism-and-self-improvement-e3f5ea64a6ac">The Craftsperson’s Toolkit: Professionalism, Pragmatism, and Self-Improvement</a> was originally published in <a href="https://medium.com/philips-technology-blog">Philips Technology Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[C++ Packages in 2024]]></title>
            <link>https://medium.com/philips-technology-blog/c-packages-in-2024-179ab0baf9ab?source=rss-bb923c920c2e------2</link>
            <guid isPermaLink="false">https://medium.com/p/179ab0baf9ab</guid>
            <category><![CDATA[software]]></category>
            <category><![CDATA[c-plus-plus-language]]></category>
            <category><![CDATA[software-development]]></category>
            <dc:creator><![CDATA[Philips]]></dc:creator>
            <pubDate>Fri, 18 Oct 2024 09:38:03 GMT</pubDate>
            <atom:updated>2024-10-18T09:38:03.528Z</atom:updated>
            <content:encoded><![CDATA[<h3>Insights from Philips Radiology Informatics</h3><p>Author: Nadav Sinai, Radiology Informatics R&amp;D Technical Leader</p><p>When creating products and solutions for healthcare, customers expect you to be able to respond quickly to bugs, security flaws, and requests for new features. Quick response is needed in any domain, to react to the market or to stay ahead of the competition, but In healthcare, a swift response is even more essential, for what’s at stake is the ability to keep hospitals running productively and maintain the safety of patients’ personal and health data.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/600/1*M7v3dtwUp7NPd8604t_Cww.jpeg" /></figure><p>Healthcare software developers need a solid, manageable, and secure codebase to meet these goals. At Philips, we take these requirements seriously, as our products serve hundreds of millions of people across the globe.</p><p>Many of our products are built with C++, which remains one of the top choices of programming languages for embedded systems and software that interface with hardware. In recent years, the</p><p>consumption and packaging of C++ libraries has been changing. However, as with many things in the world of C++, such changes occur slowly. This is in conflict with our goal of serving our customers as quickly and efficiently as possible, so we decided to address it and created a reliable software package management pipeline for our C++ systems.</p><p>In this blog post, I will describe the landscape I encountered while trying to move third-party code into packages within the Philips Vue PACS system, which has been developed in C++ for over 30 years.</p><p>Background to C++ Packaging</p><p>Unlike many modern languages such as Go and Rust, where packaging standards are built into the language or natural toolset, as of 2024, there still isn’t a standard tool or central package repository for C++. Furthermore, unlike non-native languages that compile to a known runtime platform, sharing source code in C++ is not trivial. C++ projects are closely tied to the compilation toolchains used to develop them. Library authors often cannot offer pre-compiled binaries for all platforms due to the intricate compilation process, including CPU-specific optimizations, static/dynamic linking, and changes for library-specific features.</p><p>Compatibility with the application binary interface (ABI) compatibility is essential when linking a library to a project’s binary. The ABI is primarily based on the OS/platform, which sets standards such as calling conventions and exception handling. But in practice, compilation with a different compiler or even just an upgrade to the compiler version can cause linking errors or runtime bugs. Therefore, for the best compatibility, the library code should be compiled on the same platform and preferably using the same build chain as the application.</p><p>My background is in web development with Typescript and C#, where the packaging story is clear. To understand how developers manage their third-party dependencies, I referred to two surveys: <a href="https://isocpp.org/blog/2023/04/results-summary-2023-annual-cpp-developer-survey-lite">the ISO C++ survey of 2023</a> and the <a href="https://www.jetbrains.com/lp/devecosystem-2023/cpp/#cpp_dependencymanager_two_years">JetBrains C++ ecosystem survey.</a></p><p>Below is the data on how developers manage third-party libraries:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/311/1*PpdxiwtOLAoTgubSJrgijw.jpeg" /><figcaption>Source: ISO C++ survey</figcaption></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/273/1*XLfDG_GPDfIfXGWgp2CoNg.png" /><figcaption>Source: ISO C++ survey</figcaption></figure><p>In both surveys, developers indicated that the most frustrating aspect of C++ development is managing third-party code. As I have already pointed out, C++ was not created with source-code portability in mind. The code reuse and OSS mentality that is so common today was not the norm in the 70s and 80s.</p><p>In my view, some of the answers in the surveys are practically the same — for if you’re taking in library source files into your project or prebuild the binaries per the library’s instructions, the manual work involved differs for each library, and much of that effort would probably have to be re-invested if one would need to upgrade the library version, as those instructions may change.</p><p>Also, having integrated library source code brings in the danger of it being changed (intentionally or accidentally) inside the consuming project and thus making upgrades harder.</p><p>Downloading prebuilt binaries from library repositories and sites is not much different either. It’s manual work, and the available builds may not always meet specific requirements, especially when there are transitive dependencies — e.g., ACE prebuilt with a different SSL version to the one you’re using, etc.</p><p>In any of these setups, it can be difficult to control which version of a dependency lib the prebuilt lib was compiled with. For example, say that you sourced ACE from apt, and ACE uses OpenSSL — but which version was it referring to during compile time? And if I need to upgrade to OpenSSL -will ACE link (and work) with the updated version?</p><p>A major challenge in packaging systems is known as “The Diamond Problem”: if lib A uses lib C 1.x, and lib B uses Lib C 2.x, which version of lib C should be used?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/629/1*AmCGHKGo1PX3U5mqzj4udg.jpeg" /></figure><p>In my experience, the only valid option for addressing third-party packages in a project of any meaningful size is to use package managers, which can be system-level or project-level.</p><p>If you need more than one version of a lib or declarative restoration of the dependencies, you will require a project-level package manager.</p><p>Some projects will use system-level package managers since their distribution platform is in a sandboxed container, where a Dockerfile becomes the declarative source of truth of the project</p><p>dependencies. Note that this “declarative-ness” is not so useful for automations such as Dependabot compared to declarative manifests that project-based package managers use.</p><p>In our case, we use Windows-based distribution, without containers. And since we have many supported versions — each with its own set of lib versions — we needed a declarative project-level package manager that does not “pollute” the environment.</p><p>C++ in Philips Vue PACS</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/348/1*O4hLt_KGqFveopXT_gaE3Q.jpeg" /><figcaption>Source : Philips Radiology Informatics</figcaption></figure><p>Radiology informatics PACS RCD has many products, most of which are composed of a single monorepo with 12M lines of code, 2.6M of which are in C/C++. We rely on numerous third-party libraries, ranging from common low-level abstractions such as Boost and ACE to specifics of Philips proprietary DICOM fields.</p><p>Our third-party folder contains over 100 libraries, each with its own compilation story. Dealing with the variance in our third-party dependencies was challenging, especially when addressing CVE security issues and performing upgrades. Furthermore, keeping large binaries in Git caused performance issues, which we attempted to mitigate using Git LFS and submodules. However, this caused other complications, particularly when libraries used other open-source libraries.</p><p>Considering C++ Package Managers</p><p>As I began evaluating package managers, I sought the following features:</p><p>- Declarative configuration of dependencies</p><p>- Seamless integration with Visual Studio</p><p>- Full control over the sourcing, compiling, and post-processing of packages.</p><p>- Reliable performance — once lib compilation is done for a set ABI — it will not happen again — binary caching will allow the reuse of jobs already done.</p><p>Based on the surveys — I explored two project-level package managers -VCPKG and Conan. CONAN</p><p>Conan is an open-source project with active community support. It is rooted in the Python ecosystem and is installed using pip. The Conanfile, the “recipe” for building a package, is to be written in Python, although it also has first-class support for integrating with existing CMake-based builds.</p><p><a href="https://conan.io/">Conan.io</a> serves as a public central registry for packages, but users can also set up their own registry via Artifactory.</p><p>You can easily consume packages from Conan. You only need a conanfile.txt to specify your requirements without any Python code. But you’ll soon find that you need more control and will end up with a more intricate conanfile.py to control the static/dynamic or lib features you need.</p><p>In Feb 2023, the Conan team released a <a href="https://blog.conan.io/2023/02/22/Conan-2.0.html">new and rearchitected v2.0</a> of the package manager. So, I must say I may have looked at it differently if I did this today, but back then, I encountered a few problems with Conan on Windows C MSBuild when trying to consume ACE C SSL, which did not work well with each other. It may have been problems specific to those libraries we set for our POC.</p><p>VCPKG</p><p><a href="https://vcpkg.io/">VCPKG</a> is a C++ cross-platform package manager from Microsoft, and its roots are as a system- level package manager. However, it has added support for project-level (manifest mode) package management since 2021.</p><p>VCPKG builds on CMake. You describe packages compilation steps as you do in CMake but also have access to VCPKG-supplied helper functions to download, build, and process files.</p><p>As this is a Microsoft product, its integration with Visual Studio is exceptional.</p><p>VCPKG uses CI automation to ensure that each pull request to update a library portfile (the recipe to build a package) not only passes its own build, but also verifies that all library dependents still work with the updated library. This validation ensures that all the libs declared in a vcpkg.json configuration file have been verified to work with each other as the file also includes the registry “baseline” (git SHA) from which all dependencies portfiles are sourced.</p><p>This effectively resolves the “Diamond problem” nicely, yet VCPKG does allow for version override control.</p><p>Additionally, VCPKG features advanced binary caching capabilities, which allow you to control and debug the elements (compiler versions, environment variables, etc.) which go into making the package checksum for ABI to make sure that once we built for a certain ABI, we don’t need to build again.</p><p>There are many options for backend storage and recipes for CI integration, such as Github Actions with the use of its Action Cache for binary storage or Azure Repos pipelines and other backing stores such as Artifactory, AWS S3, etc.</p><h3>Implementation of our vcpkg integration</h3><p>Philips Vue (Picture Archiving and Communication System) PACS is an image-management platform that provides scalable local and wide-area PACS solutions for hospitals and related institutions. PACS is used primarily in healthcare organizations to securely store and digitally transmit electronic images and clinically-relevant reports. Detecting and fixing security flaws is very important to ensure the integrity of PACS systems.</p><p>In our efforts to manage third-party dependencies more effectively within the Philips Vue PACS system, we adopted VCPKG because of its strong integration with Visual Studio, ability to handle complex dependencies, and advanced binary caching capabilities.</p><p>To implement VCPKG, we created two repositories:</p><p><strong>ri-vcpkg-fork</strong>: A fork of the main VCPKG registry repository.</p><p><strong>ri-vcpkg-registry</strong>: A repository where we develop and maintain our own packages that are not available in the primary VCPKG repository.</p><p>We integrated VCPKG into our Visual Studio toolchain by configuring it to automatically include the necessary dependencies for all projects. This setup ensures that the correct versions of libraries are used consistently across different projects, reducing the risk of compatibility issues.</p><p>One of the key features we leveraged is VCPKG’s “binary caching.” This feature allows us to reuse precompiled binaries, which significantly speeds up the build process and ensures that once a library is built for a specific configuration, it doesn’t need to be rebuilt unless there are changes. This approach also helps maintain a clean and efficient development environment.</p><h3>Publishing your own packages</h3><p>To publish our own packages using VCPKG, we follow a structured process that involves creating specific configuration files. These files define how to obtain the source code, compile it, and where to place the resulting binaries and headers.</p><p>The process involves:</p><p><strong>Defining the Source: </strong>Specifying where the source code for the package can be found, typically from a version control system like Git.</p><p><strong>Compilation Instructions: </strong>Providing instructions on how to compile the source code, including any configurations for different build environments (e.g., debug or release modes).</p><p><strong>Post-Processing: </strong>Indicating additional steps needed after compilation, such as copying files to specific directories.</p><p>Once these steps are completed, the package will be ready for use on our projects. However, manually managing these configurations can be cumbersome, so we plan to automate this process to ensure consistency and reduce potential errors.</p><p><strong>Conclusion</strong></p><p>Using a package manager for C++ offers numerous advantages, such as simplifying the management of third-party code, streamlining the updating of libraries and introducing new ones. Additionally, it allows for the use of tooling for security auditing and software bill of materials (sBOM) generation. VCPKG is an excellent choice for such tasks, as it provides a comprehensive public registry that ensures the compatibility of packages. Its declarative manifest simplifies tooling integration and offers efficient and reproducible builds through its binary caching feature.</p><p>The adoption of VCPKG at Philips Radiology Informatics RCD has enabled us to tackle the challenges of managing third-party dependencies, improve our development workflow, and streamline the packaging process within the Philips Vue PACS system. By leveraging VCPKG’s advanced features and seamless integration with Visual Studio, we have simplified the management of third-party libraries and ensured efficient and reliable builds while enhancing the security and maintainability of our software.</p><p>As we prepare to deploy the new version of Vue PACS in production, we are better prepared to iterate quickly and make sure we can respond to the feedback we receive from our customers and the wider health-tech community.</p><p><strong><em>Curious about working in tech at Philips? </em></strong><a href="https://www.careers.philips.com/global/en/philips-tech-careers"><strong><em>Find out more here</em></strong></a></p><p>References:</p><p><a href="https://www.jetbrains.com/lp/devecosystem-2023/cpp/">JetBrains Survey 2023</a></p><p><a href="https://isocpp.org/files/papers/CppDevSurvey-2023-summary.pdf">ISO C++ Survey 2023</a></p><p><a href="https://vcpkg.readthedocs.io/en/latest/users/binarycaching/">https://learn.microsoft.com/en-us/vcpkg/</a></p><p><a href="https://vcpkg.readthedocs.io/en/latest/users/binarycaching/">https://conan.io/</a> Recommended reads:</p><p><a href="https://decovar.dev/blog/2022/10/30/cpp-dependencies-with-vcpkg">https://decovar.dev/blog/2022/10/30/cpp-dependencies-with-vcpkg</a></p><p><a href="https://moderncppdevops.com/pkg-mngr-roundup/">https://moderncppdevops.com/pkg-mngr-roundup/</a></p><p><a href="https://learn.microsoft.com/en-us/vcpkg/maintainers/variables">https://cloudsmith.com/blog/why-programmers-need-a-c-package-manager</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=179ab0baf9ab" width="1" height="1" alt=""><hr><p><a href="https://medium.com/philips-technology-blog/c-packages-in-2024-179ab0baf9ab">C++ Packages in 2024</a> was originally published in <a href="https://medium.com/philips-technology-blog">Philips Technology Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Navigating the Future: Philips and the Evolving AI Regulation Landscape]]></title>
            <link>https://medium.com/philips-technology-blog/navigating-the-future-philips-and-the-evolving-ai-regulation-landscape-5ff05e70bd3d?source=rss-bb923c920c2e------2</link>
            <guid isPermaLink="false">https://medium.com/p/5ff05e70bd3d</guid>
            <category><![CDATA[artificial-intelligence]]></category>
            <category><![CDATA[ai-regulation]]></category>
            <category><![CDATA[future-of-ai]]></category>
            <category><![CDATA[patient-safety]]></category>
            <category><![CDATA[responsible-ai]]></category>
            <dc:creator><![CDATA[Philips]]></dc:creator>
            <pubDate>Thu, 23 May 2024 12:28:20 GMT</pubDate>
            <atom:updated>2024-05-23T12:28:20.346Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Authors: Deha Kuran, Arlette van Wissen, Michael van Hartskamp, Ger Janssen, Philips Responsible AI Team</em></p><p>We are witnessing an exciting moment in time where artificial intelligence is becoming an increasingly mainstream technology and is being integrated into our day-to-day lives to address challenges in healthcare. As a company we have been leveraging artificial intelligence (AI), including machine learning (ML), in various applications over the past 10+ years. Hence, the question: how are we regulating this superpower to prevent misuse and maximize the benefits?</p><p>In the Philips Data &amp; AI Excellence team, we focus on enabling Philips’ businesses to harness data and AI capabilities effectively in our products, with a focus on patient safety and quality. What this enablement entails varies based on the nature of each business’ needs. These needs are dynamic and change with the regulatory landscape as well as with the rise of innovative technologies, like generative AI. Our team supports teams with questions concerning people upskilling, governance and process definitions. What these efforts have in common is making sure our AI-enabled technology is safe, complies with regulations, meets ethical standards, and is trustworthy.</p><p>In this blog post, we give an overview about the AI regulatory landscape from our line of sight. We also show its impact on healthcare technology development, and how Philips is navigating these challenges.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*OGoVE6St-4CDO6A5VC8NFg.png" /></figure><h4><strong>AI Regulations Landscape: A Philips Perspective</strong></h4><p>In response to the growing need to responsibly manage AI in healthcare technologies, there are several initiatives to bring forward principles and best practices. Our CEO Roy Jakobs is co-chair of the National Academy of Medicine’s (NAM) Steering Committee which is developing an AI Code of Conduct to ensure equitable, ethical, safe and responsible use of AI in health. They recently published their proposed <a href="https://nam.edu/artificial-intelligence-in-health-health-care-and-biomedical-science-an-ai-code-of-conduct-principles-and-commitments-discussion-draft/">AI Code of Conduct commitments:</a></p><p>1. Protect and advance human health and human connection as a primary aim.</p><p>2. Ensure equitable distribution of benefit and risk for all.</p><p>3. Engage people as partners with agency in every stage of the life cycle.</p><p>4. Renew the moral well-being and sense of shared purpose to the health care workforce.</p><p>5. Monitor and openly and comprehensibly share methods and evidence of AI’s performance and impact on health and safety.</p><p>6. Innovate, adopt, collaboratively learn, continuously improve, and advance the standard of clinical practice.</p><p>The intention of the NAM Steering Committee is that all decisions associated with, and actions taken, to develop and deploy AI in the health sector will be consistent with these commitments to develop and foster trust.</p><p>Similar commitments are already clearly visible in proposals from regulatory bodies. For instance, the US Food and Drug Administration (FDA) is paying close attention to AI-enabled medical devices and has released <a href="https://www.fda.gov/medical-devices/software-medical-device-samd/good-machine-learning-practice-medical-device-development-guiding-principles">guiding principles on good machine learning practices </a>for medical device development.</p><p>In Europe, <a href="https://digital-strategy.ec.europa.eu/en/policies/data-act">the Data Act </a>and <a href="https://digital-strategy.ec.europa.eu/en/policies/regulatory-framework-ai">AI Act</a> are set to revolutionize how AI is leveraged in the European Union (EU). The AI act categorizes AI systems based on risk — in healthcare, high-risk are those systems integral to patient care and outcomes. These high-risk systems will have to comply with strict requirements and conformity assessments. This includes extensive documentation, transparency in algorithms, and robust data governance to ensure traceability and accountability.</p><p>In China, the National Medical Products Administration (NMPA) has established <a href="https://chinameddevice.com/digital-health-nmpa-ai/#:~:text=Outline%20of%20the%20Digital%20Health%20NMPA%20AI%20Guideline%3A,databases%208%20%288%29%20White%20box%20algorithm%20More%20items">guidelines</a> that emphasize the reliability, safety, and effectiveness of AI medical devices. The NMPA’s standards mandate clinical evaluation reports, risk management throughout the product lifecycle, and post-market surveillance to safeguard public health.</p><p>These are just a few examples of established standards &amp; regulations that serve as benchmarks for various nations, many of which are in the process of developing their own policies. This is reflective of a global consensus on the need for stringent regulatory oversight to manage AI’s profound implications on healthcare.</p><h4><strong>The challenges of developing trustworthy AI systems</strong></h4><p>As you can see, the global AI regulation landscape is complex and varied. Philips remains proactive in navigating this landscape, ensuring our healthcare solutions are safe by meeting these high standards of compliance and safety. As a global healthcare company, we must invest in AI systems that not only excel in performance, but that are also created and deployed following responsible AI practices.</p><p>The unprecedented power of AI needs additional regulations, certainly with the introduction of generative AI (general purpose AI). The absence of such regulations can lead to significant issues such as biased and discriminatory outcomes, fabricated conclusions, hallucinations (failures/errors), and privacy or security concerns that may lead to unfair and unethical use.</p><p>For example, <a href="https://www.nature.com/articles/d41586-019-03228-6">a clinical study</a> found that a software widely used in the healthcare industry prioritized less sick white patients over black patients who were more severely ill for access to high-risk healthcare management programs. The bias was due to the algorithm’s reliance on healthcare costs to determine patient ‘risk’ rather than actual illness, thereby inadvertently reflecting and perpetuating existing racial disparities in healthcare spending and access.</p><p>Generative AI also has biases, which show themselves when you generate text or images. A <a href="https://www.nature.com/articles/d41586-024-00291-6">recent paper</a> has showed that men are starkly over-represented in online images, and that gender bias is stronger in generated images compared with generated text. For instance, if you generate a photo of a surgeon, you get mostly white males.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0VhPp1OHMGDNh9uzDP5P8Q.png" /><figcaption><a href="https://pubmed.ncbi.nlm.nih.gov/37966807/">source</a></figcaption></figure><p>A more recent non-healthcare example from Google’s generative AI, Gemini, shows that trying to mitigate a known bias might prove harder than you think. Gemini tried to overcome a bias of non-diversity in images by prioritizing diversity in the output. As a consequence however, when Gemini was prompted to generate images of historical figures such as ‘a Viking’ or ‘a pope’, it ended up creating diverse but historically inaccurate images. So detecting and mitigating bias is very important, but not so easy.</p><p>This a challenge that not only Google, who has <a href="https://medium.com/r?url=https%3A%2F%2Fblog.google%2Fproducts%2Fgemini%2Fgemini-image-generation-issue%2F">publicly acknowledged</a> the issue, is facing. This is a problem that everyone working on the process of developing AI-enabled solutions can encounter. Sadly, the internet is a vast landscape with lots of human bias. Despite all good intentions, supervised and unsupervised models can be trained with this biased data and can be developed with their own (implicit) biases.</p><h4><strong>Philips’ Approach to AI Challenges</strong></h4><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*kYNUQUikuyZ4Z_DD2s9j6A.png" /><figcaption>Guiding principles for AI at Philips</figcaption></figure><p>At Philips, we have a set of company-wide <a href="https://www.philips.com/a-w/about/philips-data-principles.html">Data</a> and <a href="https://www.philips.com/a-w/about/artificial-intelligence/philips-ai-principles.html">AI</a> principles that guide us in creating safe, compliant, ethical and trustworthy solutions. They emphasize our priorities when we design, develop, deploy and monitor AI. Our commitment to these principles underpins our technology development as it is our intention to ensure that all our healthcare solutions are safe and beneficial for all stakeholders.</p><p>To make these guiding principles more actionable and relevant, our Data &amp; AI Innovation Excellence team collaborates with subject matter experts and works with Philips businesses to develop a Responsible AI framework, which aims to give guidance on technical (can we?), compliance (are we allowed?) and ethical (should we?) elements of AI deployment. Below, we give a couple of examples on how we move from principles to practice.</p><p>To improve awareness of fairness considerations and mitigation of biases, we created the so-called BIRIA (Bias Risk Assessment) tool. It is being introduced into the product development lifecycle to guide data scientists and AI developers in checking where biases might occur so that the right preventive or mitigative actions can be taken by product development teams.</p><p>Another example can be given to address accountability and transparency. Internally, we are using data and algorithm catalogues to clearly show who is accountable for the data and the implemented models. In addition, we use tools to trace back the lineage of each asset and see all its attributes with any transformation happening on these data elements.</p><p>One other concrete example concerns sustainability. We focus on identifying metrics and best practices for green AI, and on methods for monitoring the environmental impact of our AI-enabled medical solutions. As you might know, training an AI model consumes a lot of power, and using AI in the field does too. We aim to carefully assess our solutions and use the metrics to drive green software and AI development, thereby striving to reduce both our environmental impact and that of our partners.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*AiqNs4aYrBCrrM_v7w0ylw.png" /></figure><p>Our Responsible AI framework is designed and being developed with flexibility to address the evolving landscape of AI regulation while putting patient safety and quality at the center of everything that we want to introduce.</p><p><strong><em>Curious about working in tech at Philips? </em></strong><a href="https://www.careers.philips.com/global/en/philips-tech-careers"><strong><em>Find out more here</em></strong></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5ff05e70bd3d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/philips-technology-blog/navigating-the-future-philips-and-the-evolving-ai-regulation-landscape-5ff05e70bd3d">Navigating the Future: Philips and the Evolving AI Regulation Landscape</a> was originally published in <a href="https://medium.com/philips-technology-blog">Philips Technology Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[How can you engineer innovations for maximum customer value and business impact?]]></title>
            <link>https://medium.com/philips-technology-blog/how-can-you-engineer-innovations-for-maximum-customer-value-and-business-impact-277ad6d61570?source=rss-bb923c920c2e------2</link>
            <guid isPermaLink="false">https://medium.com/p/277ad6d61570</guid>
            <category><![CDATA[software-development]]></category>
            <category><![CDATA[problem-solving]]></category>
            <category><![CDATA[healthcare-innovations]]></category>
            <category><![CDATA[innovation-software]]></category>
            <category><![CDATA[innovation-engineer]]></category>
            <dc:creator><![CDATA[Philips]]></dc:creator>
            <pubDate>Fri, 29 Mar 2024 13:19:57 GMT</pubDate>
            <atom:updated>2024-05-28T19:49:07.279Z</atom:updated>
            <content:encoded><![CDATA[<p><em>Author: Kishore Chivukula, Philips Senior Systems Architect</em></p><p>Innovation is one of the cornerstones of staying relevant and competitive for a business while ensuring you continue to deliver value for your customers. For healthcare in particular, the success of any innovation is determined by the fine balance between problem-solving, customer adaptation, business impact and technology choices.</p><p>But how can you make sure that your organization is making the right innovations: those that matter and have an impact?</p><p>What does it take to deliver innovations at scale, at speed, and on budget?</p><p>In my ten years of experience at Philips (this is my second tenure) as an engineer and a member of multiple engineering teams, I helped deliver innovations in platform development, cloud, and serviceability. It’s here that I learned that ‘engineering innovations’ is a discipline that helps bring ideas to market through a holistic view of the customer, business, and technology.</p><p>What I’ve learned is there are three pillars that help to build a culture that fosters innovation.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*0qAeJ-LCSuxu96AW0-2Mxw.png" /></figure><h4><strong>Pillars of an innovation culture:</strong></h4><p><strong>Information sharing</strong></p><p>○ Ensure information flows seamlessly to all levels about customers, business challenges, competitors, and, most importantly, what is and isn’t working.</p><p>○ Based on this information, engineers can relate to any problem that comes to them and know how to find the best approach to navigate it.</p><p><strong>Customer interactions</strong></p><p>○ Allow every engineer to listen to the voices of key customers regularly.</p><p>○ Encourage them to silently listen-in on customer interviews, participate in user acceptance testing (UAT) sessions or meet unhappy customers.</p><p>○ See how they come back with new roadmap items that can be the key differentiator for products and services.</p><p><strong>Platform to collaborate and innovate</strong></p><p>○ Organize hackathons, developer events, and brainstorming workshops.</p><p>○ Encourage networking sessions where teams can learn from each other through coffee or lunchtime discussions, enabling them to understand each other’s challenges and see if there are synergies for collaboration.</p><p>○ Create opportunities that allow engineers to solve challenges in multiple ways and finally pick the one that delivers the best solution.</p><p>Engineers are typically assigned to solve technical challenges. Notably, the most successful organizations are those which adopt a culture where engineers are asked to work on solving customer and business challenges using technology.</p><p>Let me make it clearer by sharing a high-level view of three innovations in which I took part.</p><p>These innovations cover three different categories:</p><p>● Business challenges</p><p>● UX challenges</p><p>● Technical challenges</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*-9T6hupy-Wv-OS_m4zCj3A.png" /></figure><p><strong>Business challenges</strong></p><p>An Architect I worked with visited one of our factories in the Netherlands to understand the pain points in the system configuration for one of our core products. In the integration phase, a factory engineer typically takes time to follow manual steps, integrate, and test the system. Any mistake in configuration causes repetitive work, which costs the factory time and money. The factory asked us to simplify the process and make it foolproof to support all kinds of configurations that would deliver a new product line efficiently.</p><p>Our team ideated this first as a hackathon idea. We automated the entire process from the customer’s SAP order to system configuration in one click by creating a factory tool and deploying it to the production floor. This innovation saved the factory ~2400 hours a year while helping Philips deliver more products and serve more customers.</p><p><strong>UX challenge</strong></p><p>Hospitals need healthcare systems with high availability to treat patients and save lives. Philips service engineers use the Log Viewer tool to quickly diagnose any field problem in case the system ever has a breakdown, apply fixes and bring the system back online for the customer. Performance issues are reported with the Log Viewer 1.0 when loading a file of monthly data. The viewer tool didn’t support lazy loading and couldn’t load multiple files, making it challenging for engineers to resolve issues on time.</p><p>Our team created a solution to load the logs efficiently using chunk load, DB indexing, customizations, and added breakthrough capability to predict problems using field knowledge. Redesigning the backend service used by this tool was my Greenbelt project and helped me understand the value of ideating in a structured manner. The final solution helped engineers save 15 minutes on each log file, which would have been the time spent to identify the root cause and decide the action needed to address the customer’s challenge.</p><p><strong>Technical challenge</strong></p><p>For development of one of our flagship next generation serviceability platforms, the business decided to use existing field service software from the previous platform to meet market commitments and reduce engineering costs (minimum of three person-years).</p><p>The new platform and existing software used different technologies (C++, COM, Web and .Net) and had to work on a medical system flawlessly. Our team took it as a challenge, developed a proof of concept using WIN32 API and Automation to finalize the approach. Reliable integration and successful delivery were achieved in less than 9 months.</p><p>I chose the above three examples for their uniqueness and impact. But as you can imagine, teams at Philips create plenty of innovations every year, and I am lucky to be part of such teams where you learn something new every week.</p><p>Through my experience, I’ve learned that expanding your skills to pick the right customer challenge, associating business value and identifying suitable technology will ultimately help you as an engineer to deliver successful innovations.</p><p><strong><em>Curious about working in tech at Philips? </em></strong><a href="https://www.careers.philips.com/global/en/philips-tech-careers"><strong><em>Find out more here</em></strong></a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=277ad6d61570" width="1" height="1" alt=""><hr><p><a href="https://medium.com/philips-technology-blog/how-can-you-engineer-innovations-for-maximum-customer-value-and-business-impact-277ad6d61570">How can you engineer innovations for maximum customer value and business impact?</a> was originally published in <a href="https://medium.com/philips-technology-blog">Philips Technology Blog</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>