<?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 Surabhi Rathore on Medium]]></title>
        <description><![CDATA[Stories by Surabhi Rathore on Medium]]></description>
        <link>https://medium.com/@surabhi.rathore03?source=rss-0e7323b74e67------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*57_YvT6sv19O9TdoeMWzJw.jpeg</url>
            <title>Stories by Surabhi Rathore on Medium</title>
            <link>https://medium.com/@surabhi.rathore03?source=rss-0e7323b74e67------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 04:54:25 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@surabhi.rathore03/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[Challenges in cardiovascular care: a personalized approach]]></title>
            <link>https://medium.com/@surabhi.rathore03/challenges-in-cardiovascular-care-a-personalized-approach-828fe7ccaffb?source=rss-0e7323b74e67------2</link>
            <guid isPermaLink="false">https://medium.com/p/828fe7ccaffb</guid>
            <dc:creator><![CDATA[Surabhi Rathore]]></dc:creator>
            <pubDate>Fri, 02 Jan 2026 22:42:08 GMT</pubDate>
            <atom:updated>2026-01-02T22:42:08.247Z</atom:updated>
            <content:encoded><![CDATA[<p>Cardiovascular disease remains a leading cause of mortality worldwide. Despite substantial advances in diagnostic imaging, treatments, and devices, there is still a pressing need for more patient-specific, predictive, and adaptive approaches to improve outcomes [1]. Real-time cardiovascular simulations represent a promising avenue for delivering personalized treatment planning, surgical guidance, device design, and continuous patient monitoring. Achieving this, however, is a considerable feat. The complexity and variability of cardiovascular physiology and the computational demands of high-fidelity simulations pose significant technical challenges.</p><h4>Challenges in real-time cardiovascular simulations:</h4><ul><li><em>Acquiring high-quality imaging data</em> (e.g., CT and MRI) and accurately segmenting anatomical structures is crucial.</li><li><a href="https://www.sciencedirect.com/science/article/abs/pii/S0045782505002586"><em>Defining appropriate boundary conditions</em></a><em> [2]</em> is crucial for accurate simulations.</li><li>Complex cardiovascular models demand <em>significant computational resources.</em></li><li>Real-time simulations require <a href="https://link.springer.com/article/10.1007/s13160-021-00469-9"><em>efficient algorithms</em></a><em> [3] </em>and<em> high-performance computing.</em></li><li><em>Computing patient-specific haemodynamics [5] </em>is challenging, especially when tailoring clinical data, such as 4D MRI data.</li></ul><p>By addressing these technical hurdles — <em>improved imaging, </em><a href="https://www.sciencedirect.com/science/article/abs/pii/S0045782505002586"><em>boundary condition specification</em></a><em>, computational efficiency, and </em><a href="https://arxiv.org/pdf/2410.20828"><em>data assimilation</em></a> — researchers and clinicians can advance real-time cardiovascular simulations, ultimately enhancing patient care.</p><h4>State-of-art for real-time cardiovascular simulations:</h4><p>State-of-the-art cardiovascular simulations rely on accurate, patient-specific outflow boundary conditions — often determined via optimal control methods — to achieve realistic flow patterns. As simulations remain computationally costly, reduced-order modelling techniques compress the underlying mathematical complexity — extracting essential features from the high-fidelity model — thereby dramatically lowering computational costs without compromising accuracy. This integration of advanced techniques enables patient-specific haemodynamic predictions, informing personalized medical decisions, such as treatment planning, device design, and surgical interventions.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/875/0*QA6NHGqT1w4htR0y.png" /></figure><h4>Optimal control problems:</h4><p>Outflow boundary conditions play a critical role in determining the accuracy and stability of cardiovascular simulations. Optimal control techniques are employed to refine outflow boundary conditions, ensuring accurate and stable cardiovascular simulations. By treating these boundary conditions as control variables and minimizing an objective function, such as the difference between simulated and measured flow velocities or pressures, researchers can achieve more realistic flow patterns <em>(patient-specific measurements obtained from 4D-flow MRI)</em> and enhance the predictive capabilities of cardiovascular flows.</p><p>Optimal control problems for cardiovascular flow involve three key components:</p><ol><li><strong>Forward model:</strong> Solving the <em>unsteady Navier–Stokes equations with suitable conditions</em> to compute haemodynamics.</li><li><strong>Cost functional:</strong> Define an objective that <em>measures the discrepancy between the simulated and target</em> flow fields.</li><li><strong>Lagrange formalism:</strong> Obtain the <em>adjoint equations to compute the gradients of the cost functional</em> concerning control variables.</li></ol><h4>Reduced order modelling:</h4><p>ROM is a technique that approximates high-fidelity, large-scale computational models with significantly fewer degrees of freedom while preserving essential accuracy. Extracting the most dominant features from the system’s behaviour is especially valuable in scenarios like parameter studies and real-time simulations.</p><p>Two specific categories of ROM [4] are distinguished:</p><ol><li><strong>Intrusive methods: </strong>These methods begin with the detailed governing equations — <em>like the Navier–Stokes equations for fluid flows</em> — and identify a set of dominant modes from high-fidelity simulations. By projecting the original equations onto this reduced set of modes, we obtain a much smaller system that still captures the underlying physics. These models ensure theoretical and physical interpretability but require direct access to modify the original solver and its equations.</li><li><strong>Non-intrusive methods: </strong>Instead of working directly with the full equations, these approaches treat the original simulation code as a <strong>“black box.”</strong> Using data-driven techniques — <em>such as machine learning, regression, or interpolation</em> — the relationship between input parameters and output fields is learned from a collection of precomputed “snapshot” data. These methods are quick to implement since they don’t alter the original solver. However, their accuracy depends on the quality of the training data, and they often provide limited insight into the underlying physics.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/875/0*rEGkjseowC60r9eW.png" /></figure><p>Carefully selecting and combining these ROM approaches can produce efficient yet accurate reduced solutions suitable for real-time analysis and decision-making.</p><h4>Conclusion:</h4><p>Real-time cardiovascular simulations can revolutionize patient care by overcoming computational, accuracy, and timing barriers. With advances in imaging, optimal control, and reduced-order modelling, these tools support personalized treatments, refined device designs, and precise surgical guidance. As they evolve, clinicians gain the ability to make more informed, personalized decisions, improving outcomes and advancing global health.</p><h4>References:</h4><ol><li>Quarteroni, Alfio e Luca Formaggia. <em>“Mathematical modelling and numerical simulation of the cardiovascular system.”</em> Handbook of numerical analysis 12 (2004).</li><li>Vignon-Clementel, Irene E., et al. <em>“Outflow boundary conditions for three-dimensional finite element modeling of blood flow and pressure in arteries.”</em> CMAME 195.29–32 (2006).</li><li>Rathore, Surabhi, et al. <em>“Numerical computation of blood flow for a patient-specific hemodialysis shunt model.” </em>JJIAM 38.3 (2021).</li><li>Hesthaven, Jan S., Gianluigi Rozza, and Benjamin Stamm. <em>“Certified reduced basis methods for parametrized partial differential equations.”</em> Vol. 590. Berlin: Springer (2016).</li><li>Rathore, Surabhi, et al. <em>“Projection-based Reduced Order Modelling for Unsteady Parametrized Optimal Control Problems in 3D Cardiovascular Flows.”</em> arXiv preprint arXiv:2410.20828 (2024).</li></ol><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=828fe7ccaffb" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>