Master MVA 2024-2025 First Semester - Medical Image Analysis based on generative, geometric and biophysical models

Hervé Delingette
Equipe Epione
2004, route des Lucioles B.P. 93
06902 Sophia Antipolis Cedex
https://team.inria.fr/epione/en/
04 92 38 77 64
Herve.Delingette@inria.fr

Xavier Pennec
Equipe Epione
2004, route des Lucioles B.P. 93
06902 Sophia Antipolis Cedex
https://team.inria.fr/epione/en/
04 92 38 76 64
xavier.pennec@inria.fr

Course Objectives

The course is an advanced introduction to the mathematical and computing tools used in medical image analysis, ranging from generic image analysis such as image registration and image segmentation to more advanced techniques in statistical and mechanical modeling for computational anatomy and physiology.

Classes are given on Tuesday afternoons from 14:00 to 17:15 at ENS Saclay in room 1Z25 (bât Nord, 1er étage)

The examination consists in the group study of a research article with a report and a 10 min presentation and a multiple choice questionnaire

Liste of courses, slides, course notes and additional material (updated in real-time)

• Tuesday Oct 1, 2024, 14:00-17:15 (ENS Saclay, salle 1Z25) Introduction to Medical Image Acquisition, Image Registration [Xavier Pennec]
  • Slides on image registration
• Tuesday Oct 8, 2024, 14:00-17:15 (ENS Saclay, salle 1Z25) Riemanian Geometry and Statistics [Xavier Pennec]
  • Slides.
  • Course notes: Introduction to differential and Riemannian Geometry [To find (much) more information on Lie groups and Riemannian manifolds, I recommend the book of Jean Gallier Notes on Differential Geometry and Lie Groups]
  • Course notes: The example of rotations
  • Course notes: Intrinsic Statistics on Riemannian Manifolds: Basic Tools for Geometric Measurements [Another presentation by Tom Fletcher: Statistics on manifolds]
• Tuesday Oct 15, 2024, 14:00-17:15 (ENS Saclay salle 1Z25) Image Filtering & Segmentation [Hervé Delingette]
  • Video du cours de 2020 (in French)
  • Slides
  • GitLab for all notebooks
• Tuesday Oct 22, 2024: 14:00-17:15 (ENS Saclay, salle 1Z25) Image Segmentation based on Clustering [Hervé Delingette]
  • Slides
  • The video of the 2023 course (in English)
  • Lecture Notes on ELBO, EM and Gaussian Mixtures
  • Notebook for Gaussian Mixture Models for Brain Segmentation
  • Notebook for K-means for Brain Segmentation
  • GitLab for all notebooks
• Tuesday Nov 5, 2024: 14:00-17:15 (ENS Saclay, salle 1Z25) Analysis in the space of Covariance Matrices [Xavier Pennec]
• Slides
• Course notes: Manifold valued image processing with SPD matrices
• Tuesday Nov 12, 2024: 14:00-17:15 (ENS Saclay, salle 1Z25) Clustering and Markov Random Fields for image segmentation [Hervé Delingette]
  • Slides
  • Course video
  • Notebook for Gaussian Mixture Models for Brain Segmentation
  • Notebook for K-means for Brain Segmentation
  • Notebook for Gaussian Mixture Models for Brain Segmentation with bias field correction
• Tuesday Nov 19 2024: 14:00-17:15 (ENS Saclay, salle 1Z25) Diffeomorphic Registration and Computational Anatomy [Xavier Pennec]
  • Video of the course (in English)
  • Slides
  • Course notes: Beyond Riemannian Geometry: The affine connection setting for transformation groups
• Tuesday Nov 26 2024: 14:00-17:15 (ENS Saclay, salle 1Z25) Active Deformable Models and Shape constraints for image segmentation [Hervé Delingette]
  • Slides
  • Course video
  • GrabCut Segmentation
  • Random Walker Segmentation
• Tuesday Dec 3, 2024, 14:00-17:15 (Visio) Exam [Hervé Delingette, Xavier Pennec].
  • Zoom link : https://univ-cotedazur.zoom.us/j/83329627570?pwd=TX5uljeJhbvuzofaOfratH7TbX4VCh.1
  • ID de réunion: 833 2962 7570
  • Code secret: 407573
  • Phone number : +33 1 8699 5831 Then enter ID followed by # then code followed by #

Resources

• Online lectures and courses at College de France

Articles for the research project (last update Nov 4, 241 Mo)

Choice of research paper

• Valentine Tosel : Most Likely Separation of Intensity and Warping Effects in Image Registration (Kuhnel et al).
• Apélété Adodo SOSSOU et Raphaêl Faure Label Refinement Network from Synthetic Error Augmentation for Medical Image Segmentation
• Lola Giordani :SynthSeg: Segmentation of brain MRI scans of any contrast and resolution without retraining (Billot et al.)
• Mathis Wetterwald et Eve Delegue Trustworthy learning with (un)sure annotation for lung nodule diagnosis with CT
• Bruna Lopes Modeling the Shape of the Brain Connectome via Deep Neural Networks
• Tom Welch et Nassim Arifette Bayesian Network for Simultaneous Keyframe and Landmark Detection.
• Géraud Ilinca. Geometry-Aware Generative Autoencoders for Warped Riemannian Metric Learning and Generative Modeling on Data Manifolds.