Modelling Simulations from Medical Images and Anatomies

IIDS67582 Modelling Simulations from Medical Images and Anatomies 15 Semester 2 Postgraduate Taught Level 7 Ali Sarrami Foroushani Unit coordinator Division of Informatics, Imaging and Data Sciences 1 FHEQ level (Framework for Higher Education Qualifications) ' Masters/Integrated Masters P4 ' European Credit Transfer & Accumulation System Rating : 7.5 This unit introduces students to how medical images can be transformed into computational representations of anatomy and physiology, and how these can be used in simulation and in-silico studies.Key themes1) Foundations of image-based modelling1.1) Patient-specific vs population-specific models1.2) Verification, validation, and uncertainty1.3) Clinical and regulatory context (in-silico trials, credibility frameworks)2) Anatomical modelling from images2.1) Segmentation, surface reconstruction, and watertight mesh generation2.2) Mesh cleaning, manifoldness, and quality control3) Shape correspondence and statistical shape modelling (SSM)3.1) Landmarking, registration, and shape variation analysis3.2) Generation and validation of synthetic geometries4) Physiological signal extraction4.1) Cleaning and calibrating Doppler Ultrasound or 4D Flow MR data4.2) Flow and pressure waveform analysis4.3) Boundary condition models5) Simplified physiological simulations5.1) Lumped-parameter networks  5.2) Sensitivity analysis and parameter interpretation6) Applications and validation6.1) Device and pathology simulations6.2) Regulatory context for in-silico trials  6.3) Credibility, reproducibility, and FAIR data principles  This unit introduces students to how medical images can be transformed into computational representations of anatomy and physiology, and how these can be used in simulation and in-silico studies.Key themes1) Foundations of image-based modelling1.1) Patient-specific vs population-specific models1.2) Verification, validation, and uncertainty1.3) Clinical and regulatory context (in-silico trials, credibility frameworks)2) Anatomical modelling from images2.1) Segmentation, surface reconstruction, and watertight mesh generation2.2) Mesh cleaning, manifoldness, and quality control3) Shape correspondence and statistical shape modelling (SSM)3.1) Landmarking, registration, and shape variation analysis3.2) Generation and validation of synthetic geometries4) Physiological signal extraction4.1) Cleaning and calibrating Doppler Ultrasound or 4D Flow MR data4.2) Flow and pressure waveform analysis4.3) Boundary condition models5) Simplified physiological simulations5.1) Lumped-parameter networks  5.2) Sensitivity analysis and parameter interpretation6) Applications and validation6.1) Device and pathology simulations6.2) Regulatory context for in-silico trials  6.3) Credibility, reproducibility, and FAIR data principles  The unit aims to:1) Introduce students to the principles and workflows of computational modelling and simulation based on medical imaging data.2) Provide students with an understanding of how quantitative anatomical and physiological models are derived from medical images.3) Develop practical skills in building simulation-ready geometries, basic physiological models, and population-level shape models using open-source tools.4) Foster critical appreciation of model credibility, validation, and regulatory frameworks for in-silico research and virtual clinical trials.  Students should be able to:<ol><li>Explain the complete image-based computational modelling workflow from medical image analysis to computational modelling and simulation.</li><li>Explain key methods in segmentation, surface reconstruction, and quality control for simulation-ready geometries.</li><li>Explain statistical shape modelling and its role in population studies and synthetic anatomy generation.</li><li>Explain basic haemodynamic principles and 0D (lumped-parameter) models such as the Windkessel.</li><li>Explain the regulatory, ethical, and validation challenges involved in translating image-based models into clinical and research applications. </li></ol> Students should be able to: 6. Critically evaluate modelling and simulation workflows in research papers, identifying assumptions and limitations.7. Select and justify appropriate computational and physiological modelling techniques for specific imaging data.8. Analyse the effect of modelling assumptions using sensitivity and uncertainty analyses.9. Integrate anatomical and physiological data to design simple in-silico experiments.  Students should be able to:10. Use open-source software to reconstruct 3D geometries from medical images.11. Prepare simulation-ready meshes and basic 0D models using Python scripts.12. Extract and process flow and pressure waveforms from Doppler or 4D Flow data.13. Evaluate model credibility through basic validation, verification, and sensitivity analyses.  ILOs 6-13 are transferable.   Each weekly 3-hour session will include:a) Interactive lecture/discussion (≈90 min): Core theory with visual demonstrations and case studies (available as recorded clips on Blackboard).b) Hands-on practical (≈90 min): Guided Python notebooks using real medical imaging datasets.c) Weekly reflection or “pipeline checkpoint”: Students update their lab log with methods, results, and notes on assumptions.  <figure class="table"><table><tbody><tr><td>Assessment task</td><td>Length</td><td>Weighting within unit</td></tr><tr><td>Formative: Pipeline checkpoints</td><td>Short weekly reflections</td><td>N/A</td></tr><tr><td>Summative: Written Exam (assessed ILOs 1-9)</td><td>60 minutes</td><td>40%</td></tr><tr><td>Summative: Case Study Project (assessed ILOs 6-13)</td><td>10 minute presentation + Short report (1500 max words)</td><td>60%</td></tr></tbody></table></figure> Formative: Pipeline checkpoints (ungraded) will receive feedback at the end of the sessionSummative: Marks back within 3 weeks MSc Applied AI for Med Imaging MSc Applied AI for Med Imaging Not Set Mandatory N Books / JournalsWesterhof, Nico, Nikos Stergiopulos, and Mark IM Noble. Snapshots of hemodynamics: an aid for clinical research and graduate education. Boston, MA: Springer US, 2005.Nichols, Wilmer W., et al., eds. McDonald’s blood flow in arteries: theoretical, experimental and clinical principles. CRC press, 2022.Ottesen, Johnny T., Mette S. Olufsen, and Jesper K. Larsen. Applied mathematical models in human physiology. Society for Industrial and Applied Mathematics, 2004.Sarrami‐Foroushani, Ali, Toni Lassila, and Alejandro F. Frangi. "Virtual endovascular treatment of intracranial aneurysms: models and uncertainty." Wiley Interdisciplinary Reviews: Systems Biology and Medicine 9.4 (2017): e1385.Sarrami-Foroushani, Ali, et al. "In-silico trial of intracranial flow diverters replicates and expands insights from conventional clinical trials: Supplementary material." Websites<a href="https://www.paraview.org/tutorials/" target="_blank">https://www.paraview.org/tutorials/</a><a href="https://www.itksnap.org/docs/viewtutorial.php?chapter=TutorialSectionIntroduction " target="_blank">https://www.itksnap.org/docs/viewtutorial.php?chapter=TutorialSectionIntroduction </a> eAssessment 12 Lectures 24 Project supervision 18 96