<?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 Shruti Kunapuli on Medium]]></title>
        <description><![CDATA[Stories by Shruti Kunapuli on Medium]]></description>
        <link>https://medium.com/@shrutikp?source=rss-f17a343e6c37------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*6fZqhK5CGQxMR9hRw0qEIA.jpeg</url>
            <title>Stories by Shruti Kunapuli on Medium</title>
            <link>https://medium.com/@shrutikp?source=rss-f17a343e6c37------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 14:46:09 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@shrutikp/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[Writing Cleaner Python Code with Pylint: A Practical Guide for Developers]]></title>
            <link>https://shrutikp.medium.com/writing-cleaner-python-code-with-pylint-a-practical-guide-for-developers-eb573ea1fcd3?source=rss-f17a343e6c37------2</link>
            <guid isPermaLink="false">https://medium.com/p/eb573ea1fcd3</guid>
            <dc:creator><![CDATA[Shruti Kunapuli]]></dc:creator>
            <pubDate>Tue, 03 Mar 2026 01:39:25 GMT</pubDate>
            <atom:updated>2026-03-03T01:39:25.518Z</atom:updated>
            <content:encoded><![CDATA[<p>Most developers focus on making their Python code <em>work</em>. Fewer focus on making it <em>clean, maintainable, and professional</em>.</p><p>But as projects grow — especially in data science, machine learning, or backend systems — messy code quickly becomes a liability.</p><p>This is where <strong>Pylint</strong> becomes extremely valuable.</p><p>In this article, we’ll explore what Pylint is, why it matters, and how to use it effectively to improve your Python code quality.</p><h3>What Is Pylint?</h3><p><strong>Pylint</strong> is a static code analysis tool for Python.</p><p>“Static analysis” means it examines your code <strong>without executing it</strong>. Instead of running your program, it analyzes the structure, style, and potential logic issues directly from the source code.</p><p>Pylint checks for:</p><ul><li>Syntax errors</li><li>Undefined variables</li><li>Unused imports</li><li>Code smells</li><li>PEP 8 style violations</li><li>Refactoring opportunities</li></ul><p>It also provides a <strong>code quality score out of 10</strong>, which makes improvement measurable.</p><h3>Why Should You Care About Code Quality?</h3><p>You might think:</p><blockquote><em>“If my script runs correctly, isn’t that enough?”</em></blockquote><p>Not really.</p><p>Here’s why clean code matters:</p><h3>1. Maintainability</h3><p>Clean code is easier to update and extend.</p><h3>2. Collaboration</h3><p>Readable code helps teammates understand your logic quickly.</p><h3>3. Debugging</h3><p>Well-structured code reduces hidden bugs.</p><h3>4. Professionalism</h3><p>Code quality standards matter in job interviews and production environments.</p><p>Especially in data science workflows, notebooks can become chaotic. Pylint helps introduce discipline into your coding process.</p><h3>Installing Pylint</h3><p>Installing Pylint is straightforward:</p><pre>pip install pylint</pre><pre>To verify installation:</pre><pre>pylint --version</pre><h3>Running Pylint on Your Code</h3><p>To analyze a Python file:</p><pre>pylint my_script.py</pre><p>To analyze a folder:</p><pre>pylint my_project/</pre><pre>After execution, Pylint outputs detailed messages and a final rating.</pre><p>Example:</p><pre>my_script.py:1:0: C0114: Missing module docstring (missing-module-docstring)<br>my_script.py:5:4: W0612: Unused variable &#39;x&#39; (unused-variable)</pre><pre>Your code has been rated at 8.00/10</pre><pre>Understanding Pylint Message Types</pre><p>Each message starts with a letter:</p><p>CodeMeaningCConvention (style issue)RRefactor suggestionWWarningEErrorFFatal error</p><p>For example:</p><pre>W0612: Unused variable &#39;x&#39;</pre><p>This means you defined a variable but never used it — a common issue in beginner code.</p><h3>Common Issues Pylint Detects</h3><h3>Missing Docstrings</h3><p>Pylint encourages documentation:</p><pre>&quot;&quot;&quot;This module handles data preprocessing.&quot;&quot;&quot;</pre><pre>While small scripts may not seem to need documentation, larger projects absolutely do.</pre><h3>Unused Variables</h3><pre>x = 10</pre><pre>If x is never used, Pylint flags it. Removing unused variables makes code cleaner and prevents confusion.</pre><h3>Poor Naming Conventions</h3><p>Bad:</p><pre>X = 5</pre><pre>Better:</pre><pre>feature_count = 5</pre><p>Descriptive names improve readability significantly.</p><h3>The Pylint Score: Should You Aim for 10/10?</h3><p>Pylint gives your code a rating:</p><pre>Your code has been rated at 8.50/10</pre><pre>While a perfect 10 is satisfying, it’s not always necessary.</pre><p>The goal is not perfection — it’s <strong>clarity and maintainability</strong>.</p><p>In real-world projects, some rules may be intentionally disabled. What matters is consistent standards across your codebase.</p><h3>When Should You Disable Warnings?</h3><p>Sometimes Pylint may feel overly strict.</p><p>You can disable specific warnings:</p><pre>value = 10  # pylint: disable=invalid-name</pre><pre>Or for an entire file:</pre><pre># pylint: disable=missing-module-docstring</pre><pre>However, avoid disabling rules without understanding them first. Pylint is often pointing out a genuine improvement opportunity.</pre><h3>Pylint in Data Science and Kaggle Projects</h3><p>In competitive environments like Kaggle:</p><ul><li>Clean notebooks are easier to iterate on</li><li>Reusable functions save time</li><li>Structured code reduces experimentation errors</li><li>Sharing code publicly demands clarity</li></ul><p>Using Pylint in data science may feel uncommon, but it significantly improves long-term productivity.</p><h3>Final Thoughts</h3><p>Pylint is more than a style checker — it’s a discipline tool.</p><p>It pushes you to:</p><ul><li>Write clearer variable names</li><li>Structure functions better</li><li>Document your logic</li><li>Avoid hidden bugs</li></ul><p>At first, it may feel strict. Over time, it trains you to write better Python naturally.</p><p>If you want to move from writing “working scripts” to writing <strong>professional-quality Python</strong>, integrating Pylint into your workflow is a powerful step.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=eb573ea1fcd3" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Unraveling the Magic of Large Language Models: The Brains Behind AI]]></title>
            <link>https://shrutikp.medium.com/unraveling-the-magic-of-large-language-models-the-brains-behind-ai-5b58dc76200a?source=rss-f17a343e6c37------2</link>
            <guid isPermaLink="false">https://medium.com/p/5b58dc76200a</guid>
            <dc:creator><![CDATA[Shruti Kunapuli]]></dc:creator>
            <pubDate>Sun, 19 Jan 2025 20:04:18 GMT</pubDate>
            <atom:updated>2025-01-19T20:08:55.674Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*8uYxBhLJOVbSRLNQ" /><figcaption>Image generated using Chat-GPT</figcaption></figure><p>The rise of Large Language Models (LLMs) has sparked a revolution in artificial intelligence. From answering complex questions to creating artful prose, these models are reshaping how humans interact with machines. But what exactly are LLMs, and how do they work? Let’s take a deep dive into this fascinating technology.</p><h3>What Are Large Language Models (LLMs)?</h3><p>Large Language Models are advanced AI systems designed to understand, generate, and interact with human language. These models are trained on massive datasets comprising books, articles, and online content to mimic the intricacies of human communication.</p><p>Popular LLMs include OpenAI’s GPT series, Google’s BERT and PaLM, and Meta’s LLaMA. They excel in tasks like:</p><ul><li><strong>Text generation</strong>: Writing articles, stories, or even code.</li><li><strong>Summarization</strong>: Condensing long texts into concise summaries.</li><li><strong>Translation</strong>: Bridging language barriers with accuracy.</li><li><strong>Question-Answering</strong>: Providing meaningful responses to user queries.</li></ul><h3>The Technology Behind the Magic</h3><p>At the heart of every LLM lies the <strong>Transformer architecture</strong>, a groundbreaking model introduced in 2017 by Vaswani et al. Transformers use a mechanism called <strong>self-attention</strong>, allowing them to focus on relevant parts of a text to understand context better. Here’s how it works:</p><h3>1. Self-Attention: Finding What Matters</h3><p>Imagine you’re reading a sentence: <em>“The cat, which was chased by the dog, climbed the tree.”</em> To understand “cat,” your brain naturally pays attention to surrounding words like “chased” and “tree.” Self-attention mimics this human capability, letting LLMs understand the relationships between words.</p><h3>2. Pretraining and Fine-Tuning</h3><p>LLMs learn in two phases:</p><ul><li><strong>Pretraining</strong>: The model is exposed to vast amounts of text, learning to predict the next word in a sequence or fill in missing words.</li><li><strong>Fine-tuning</strong>: It is then specialized for tasks like sentiment analysis or chat-based interaction.</li></ul><h3>3. Scale: Bigger is Better</h3><p>The “large” in LLM refers to the number of parameters — values the model adjusts during training. GPT-3, for example, has 175 billion parameters, enabling it to grasp complex language patterns and nuances.</p><h3>Real-World Applications of LLMs</h3><p>LLMs have transcended academia and entered the mainstream. Here are a few ways they’re making waves:</p><h3>1. Personal Assistants</h3><p>Virtual assistants like ChatGPT, Siri, and Alexa use LLMs to answer queries, set reminders, and even crack jokes.</p><h3>2. Content Creation</h3><p>Writers and marketers leverage LLMs for brainstorming ideas, drafting articles, or generating engaging copy. Tools like Jasper AI are helping professionals boost productivity.</p><h3>3. Healthcare</h3><p>In medicine, LLMs assist doctors by summarizing patient notes, offering diagnostic suggestions, and even translating complex medical jargon into layman’s terms.</p><h3>4. Code Generation</h3><p>Platforms like GitHub Copilot use LLMs to suggest code snippets, debug errors, and accelerate software development.</p><h3>Challenges and Ethical Concerns</h3><p>Despite their brilliance, LLMs come with challenges:</p><h3>1. Bias in Outputs</h3><p>Because LLMs learn from data created by humans, they can inadvertently replicate societal biases. Ensuring fairness and inclusivity in AI remains a pressing issue.</p><h3>2. Hallucinations</h3><p>LLMs sometimes generate plausible but incorrect or nonsensical answers, a phenomenon known as “hallucination.”</p><h3>3. Environmental Impact</h3><p>Training large models requires significant computational resources, raising concerns about energy consumption and sustainability.</p><h3>4. Misinformation</h3><p>LLMs can generate convincing fake news or misleading content, underscoring the need for responsible usage.</p><h3>The Future of LLMs</h3><p>The journey of LLMs is far from over. Here’s what the future holds:</p><h3>1. Multimodal Models</h3><p>Next-generation models, like OpenAI’s GPT-4 Vision, are extending capabilities to include text, images, and even audio processing, making AI more versatile.</p><h3>2. Industry-Specific Models</h3><p>Specialized LLMs tailored for domains like healthcare, law, or finance are on the rise, offering precise and contextual insights.</p><h3>3. Efficiency Innovations</h3><p>Researchers are working on techniques like model distillation and sparsity to make LLMs smaller, faster, and more accessible.</p><h3>Why LLMs Matter</h3><p>Large Language Models are more than just tools — they are reshaping our relationship with technology. They empower businesses, democratize knowledge, and push the boundaries of what’s possible in AI. However, as we unlock their potential, we must also navigate the ethical and societal challenges they pose.</p><p>As LLMs evolve, one thing is clear: the age of intelligent, human-like machines has arrived, and it’s only the beginning.</p><p>What are your thoughts on LLMs? Are they a boon or a bane? Let’s discuss in the comments below!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5b58dc76200a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Fairness, Bias, and Explainability: Why They’re the Heart of Responsible AI]]></title>
            <link>https://shrutikp.medium.com/fairness-bias-and-explainability-why-theyre-the-heart-of-responsible-ai-821e80615bcd?source=rss-f17a343e6c37------2</link>
            <guid isPermaLink="false">https://medium.com/p/821e80615bcd</guid>
            <dc:creator><![CDATA[Shruti Kunapuli]]></dc:creator>
            <pubDate>Thu, 09 Jan 2025 19:40:12 GMT</pubDate>
            <atom:updated>2025-01-09T19:40:12.220Z</atom:updated>
            <content:encoded><![CDATA[<h3>Why This Matters</h3><p>Imagine this: You apply for a loan, and it’s denied. When you ask why, the answer is, “That’s just what the algorithm decided.” Not very comforting, is it?</p><p>This kind of scenario shows why <strong>fairness</strong>, <strong>bias</strong>, and <strong>explainability</strong> are essential in AI and data science. AI is shaping our lives — whether it’s screening job applicants, approving loans, or assisting doctors in diagnosing diseases. While it has the power to transform industries, it also comes with risks. If AI systems aren’t fair, they could unintentionally exclude or harm certain groups. If they aren’t explainable, they can feel like mysterious black boxes, leaving us to blindly trust — or reject — their decisions.</p><p>Let’s break this down a bit more and explore why these three principles are critical to building AI systems we can trust.</p><h3>Fairness and Bias: Making AI Work for Everyone</h3><p>When we think about fairness in AI, it’s about ensuring the system doesn’t discriminate — intentionally or unintentionally. But here’s the tricky part: AI learns from data, and data reflects the world as it is. And let’s be honest — our world isn’t always fair.</p><p>Take, for example, the story of Amazon’s AI recruiting tool. This system was trained on resumes from the past, but those resumes reflected a male-dominated tech industry. The AI started penalizing resumes with words like “women’s” (e.g., “women’s chess club”). Removing gender wasn’t enough because the system still picked up on indirect clues tied to gender.</p><p>So, how do we fix this?</p><ul><li><strong>Start with better data</strong>: Balance datasets to include diverse and representative samples.</li><li><strong>Use fairness metrics</strong>: Tools like <em>Equal Opportunity</em> ensure that outcomes (like a job interview offer) are consistent across demographic groups.</li><li><strong>Bias detection tools</strong>: AI can also help us identify where bias is creeping in, so we can take action.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*1TCRrcm0_vEXJYN1" /></figure><h3>Explainability: Let’s Not Fear the “Black Box”</h3><p>Now imagine you’re a doctor using AI to help diagnose patients. The AI flags someone as high-risk for a condition, but doesn’t explain why. Would you feel comfortable acting on that? Probably not.</p><p>This is where <strong>explainability</strong> comes in — it’s about showing how an AI made its decision. Think of it like peeling back the curtain to reveal the factors influencing the system’s choices.</p><p>One popular tool is <strong>SHAP (SHapley Additive exPlanations)</strong>, which breaks down a prediction to show how much each feature (like a patient’s age, symptoms, or test results) contributed to the final outcome. For doctors, this could mean understanding that a patient’s high blood pressure was a key factor in their risk score — not just trusting the algorithm blindly.</p><h3>Why It All Comes Together</h3><p>Fairness and explainability aren’t just nice-to-haves; they’re the foundation of trust. If people feel like an AI system is biased or can’t explain itself, they won’t use it — or worse, it could cause real harm. And this isn’t just about ethics; it’s about business and compliance too. Laws like GDPR require fairness and transparency in automated decision-making.</p><p>So, what can companies and data scientists do?</p><ol><li><strong>Involve diverse teams</strong>: Build AI systems with input from people of different backgrounds to spot blind spots early.</li><li><strong>Test, monitor, repeat</strong>: Fairness and transparency aren’t one-time fixes. Regularly audit models to catch changes or unexpected behavior.</li><li><strong>Be transparent</strong>: Communicate how your AI works in simple, clear terms that anyone can understand — not just data scientists.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*kwa1mZLF0DFbptuO.png" /></figure><h3>Final Thoughts</h3><p>At its best, AI can be a game-changer — helping us make better decisions, faster. But to truly realize its potential, we need to ensure it works for everyone, not just a select few. Fairness, bias, and explainability aren’t just technical challenges — they’re about keeping AI human-centric.</p><p>So the next time you hear about an AI system, ask: Is it fair? Is it transparent? And most importantly — can it explain itself? Because that’s how we build systems we can trust.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=821e80615bcd" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[A Gentle Introduction to AI for Medical Imaging — Translating radiographic medical images to…]]></title>
            <link>https://shrutikp.medium.com/a-gentle-introduction-to-ai-for-medical-imaging-translating-radiographic-medical-images-to-704f46c372d8?source=rss-f17a343e6c37------2</link>
            <guid isPermaLink="false">https://medium.com/p/704f46c372d8</guid>
            <category><![CDATA[machine-learning]]></category>
            <category><![CDATA[deep-learning]]></category>
            <category><![CDATA[begineers]]></category>
            <category><![CDATA[healthcare]]></category>
            <category><![CDATA[artificial-intelligence]]></category>
            <dc:creator><![CDATA[Shruti Kunapuli]]></dc:creator>
            <pubDate>Sat, 13 Feb 2021 19:01:30 GMT</pubDate>
            <atom:updated>2021-06-04T01:57:48.609Z</atom:updated>
            <content:encoded><![CDATA[<h3><strong>A Gentle Introduction to AI for Medical Imaging — Translating radiographic medical images to derive meaningful clinical endpoints</strong></h3><p>Medical Imaging plays an important role in modern medicine. It allows us to visualize and examine the human body in depth which manifests structures inside our body in great detail. Nowadays, medical devices are constantly becoming more advanced in terms of — cost, precision, and safety — to improve the medical experience of the end-users.</p><p>After the recent boost in the field of AI, effective solutions came to light to solve countless intricate problems in multiple sectors using both machine learning and deep learning algorithms.</p><p>One such area that has gained continuous momentum in the aspect of research and development is the healthcare domain.</p><p>Few reasons why AI in healthcare is flourishing day by day are as follows:-</p><ol><li>The amount of data generated from medical devices is extensive- In an era where data is produced faster than its consumed, processing such a volume of data by humans alone induces a lot of physician burnout.</li><li>Furnish high accuracy and precision- AI can enhance efficiency, reduce human errors, and achieve goals with minimal human operation.</li><li>Extremely cost-effective- developing and maintaining AI solutions are less expensive when compared to bringing in doctors from across the globe to screen a patient’s health conditions.</li></ol><blockquote>Want to read this story later? <a href="https://usejournal.com/?utm_source=medium.com&amp;utm_medium=noteworthy_blog&amp;utm_campaign=tech&amp;utm_content=guest_post_read_later_text">Save it in Journal.</a></blockquote><h3>The flow of AI in medical imaging</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/768/1*T9NTh_Vd2l4r0QMyB183hw.png" /><figcaption><em>Fig:- Real-world flow of medical imaging- 1) Imaging tools, 2) radiologist to provide insights about images, 3) Picture archiving and communication system</em>(<em>PACS) to store image and patient data, 4) Diagnostic Clinician to examine patients and render a diagnosis. AI can be incorporated in any of the steps in the flow.</em></figcaption></figure><p>Now let us dive into each segment present in the flow in order to understand the process better. Firstly, imaging tools are used to generate image data using advanced imaging techniques.</p><h4>Imaging tools</h4><p>Diagnostic imaging tools help narrow the causes of an injury or illness and ensure that the diagnosis is accurate. These techniques include X-Rays, computed tomography (CT) scans, and magnetic resonance imaging (MRI).</p><p>These imaging tools let your doctor “see” inside your body to get a “picture” of your bones, organs, muscles, tendons, nerves, and cartilage. Such that the doctor can determine if there are any abnormalities present.</p><p>Imaging tools are broadly classified into two categories: 2-Dimensional and 3-Dimensional imaging tools.</p><ul><li><strong>2D imaging</strong>: an imaging technique where pictures are taken from a single angle. Example:- X-Ray, Ultra Sound Scan, and Microscopy.</li><li><strong>3D imaging</strong>: an imaging technique where pictures are taken from different angles to create a volume of images. Example:- Computed Tomography (CT), Magnetic Resonance Imaging (MRI)</li></ul><p>The three most commonly used imaging tools include x-rays, computed tomography (CT) scans, and magnetic resonance imaging (MRI) which are differentiated in the table below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/890/0*tkIR3wwXgW38kJnY" /></figure><p>The generated data from the imaging tools are viewed by <strong><em>Radiologists</em></strong>, who are the primary readers of the image. They analyze the image and write a small note based on their observations, which is understood by a commoner without any medical knowledge. Radiologists create a specific file format called DICOM combining all the data present upon which the data is moved to a storage unit called PACS using specific file formats.</p><p><strong><em>Picture Archiving and Communication System(</em></strong>PACS<strong><em>)</em></strong></p><p>PACS Systems are a medical imaging technology that provides economical storage, retrieval, management, distribution, and presentation of medical images. Electronic images and reports are transmitted digitally via PACS systems. This eliminates the need to manually file, retrieve, or transport film jackets. It allows a healthcare organization (such as a hospital) to capture, store, view, and share all types of images both internally and externally.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*zfHhalFBdbcfrxFM" /><figcaption>Source: Wikipedia</figcaption></figure><p>Equivalent to pdfs and docs on our system folder, PACS stores files either in DICOM or NIFITI format. Digital Imaging and Communications in Medicine(DICOM) is the standard for the communication and management of medical imaging information and related da ta. DICOM files consist of both patient information along with image information and modality.</p><p>The final step in the flow is a<strong> <em>Diagnostic Clinician</em> </strong>who<strong> </strong>takes the report provided by the radiologist from the storage system along with a few other parameters like, patient’s current condition/symptoms, other labs result, and prior medical knowledge to provide the diagnosis.</p><h3>Different Machine Learning Techniques Used</h3><ol><li><strong>Classification</strong>:- Classification is a technique where we categorize data into a given number of classes. The main goal of a classification problem here is to identify whether a disease or anomaly is present in the image.</li></ol><p>Use case- Identifying the presence of Pneumonia from lung X-Rays.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/762/0*rdwNv8Pt_adfjGt2" /></figure><p>2. Segmentation:- Segmentation in Image Processing is being used in the medical industry for efficient and faster diagnosis, detecting diseases by identifying specific pixels in an image that contains a structure or finding.</p><p>Use case- Identify tumors and calculate the size of the abnormality.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/300/0*51cGo8713hIyuFE4" /><figcaption>Source: <a href="https://www.chimaera.de/">Chimaera GmbH</a> gallery</figcaption></figure><p>3. Localization:- Localization is used to narrow down the area of the image that may contain the abnormality.</p><p>Use case- Helps to draw attention to mass in breasts to identify the presence of Breast cancer.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/225/0*mRGL83WRAuWGdZ-Q" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/225/0*mFV2yev7E0jzq02Y" /><figcaption>Source: <a href="https://www.researchgate.net/publication/335926532_Localization_with_Limited_Annotation">Localization </a>research</figcaption></figure><p>The final step after building a machine learning model is to get medical approval from a certified regulatory body such as — FDA. FDA is a federal agency that is responsible for protecting public health by ensuring the safety, efficacy, and security of human and biological products. Only upon approval from the regulatory body allows us to use the model/solution to treat patients in the real-world. Which generally is a very time taking process with a lot of standards that need to be met.</p><p>AI in the healthcare industry can be used in various scenarios from disease detection to prioritization of records in PACS to fasten the retrieval process. However, any solution that is provided using algorithms irrespective of its high performance can only contribute to the betterment of the results but not to replace Radiologists/Doctors as the scope of the model is limited to a specific problem statement and will fail to take into account a holistic view of a patient’s condition.</p><p>📝 Save this story in <a href="https://usejournal.com/?utm_source=medium.com&amp;utm_medium=noteworthy_blog&amp;utm_campaign=tech&amp;utm_content=guest_post_read_later_text">Journal</a>.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=704f46c372d8" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>