Machine Learning for Medical Image Analysis (Seminar)

Course title:Machine Learning for Medical Image Analysis
Course ID:ML-4506
Lecturers:Dr. Lisa Koch, Dr. Christian Baumgartner
Teaching assistants:Paul Fischer, Indu Ilanchezian, Nikolas Morshuis, Sarah Müller, Stefano Woerner
Time:Friday 12-14h
Location:Maria-von-Linden-Straße 6 - 4th floor seminar room

The seminar starts with an introductory lecture to provide a compact overview of the research field (machine learning for medical image analysis), as well as a tutorial on critical analysis and presentation of research papers. Throughout the remainder of the course, each student presents two papers from a collection of seminal work in the field. Strong emphasis will be put on an engaging group discussion of the paper.

The learning objectives of this seminar consist of three parts: (1) the students will gain a solid understanding of key contributions to the field of machine learning for medical image analysis, (2) the students learn to critically read and analyse original research papers and judge their impact, and (3) the students will improve their scientific communication skills with an oral presentation and participation in discussions sessions.

News

  • 20.10.: Paper selection and schedule is online
  • 6.10.: We have reached full capacity! Please sign up for the waiting list though. There are usually a few people who unenroll before the semester starts, so you might still get a spot!
  • 24.9.: Registration in ILIAS is now open
  • 15.9.: Website is online

Important information

Please register for the course in ILIAS. Number of participants is limited to 12 students. We will maintain a waiting list - if you do not wish to take the course, please unenroll to let other people take your place.

Each student chooses two papers from the provided collection to present during the course of the seminar. The students may get support in the preparation of their presentation by the seminar assistants. Everybody is encouraged to read each paper before it is being presented and engage in a discussion following the presentations. To foster interesting discussions, each paper will also be assigned two “critics” who study the paper and prepare questions for the discussion. Each student will be graded based on both their presentation (80%) and their participation in the assigned discussions (20%). Attendance is required to pass the course (3 absences allowed).

Slides

Slides for the introductory lecture will be uploaded to ILIAS.

Schedule and List of Papers

Note: contact information of the TAs is provided in the course materials.

Seminar datePaper No.TitleTA
22.10.2021Introduction to course and to reading and presenting of scientific work
29.10.2021Introduction to medical image analysis
Paper 1U-Net: Convolutional Networks for Biomedical Image Segmentation (2015)Sarah
5.11.2021Paper 2nnU-Net: a self-configuring method for deep learning-based biomedical image segmentation (2021)Nikolas
Paper 3Integrating statistical prior knowledge into convolutional neural networks (2017)Lisa
12.11.2021Paper 4Test-time adaptable neural networks for robust medical image segmentation (2020)Indu
Paper 5Fairness in cardiac mr image analysis: An investigation of bias due to data imbalance in deep learning based segmentation (2021)Lisa
19.11.2021Paper 6[cancelled] Nonrigid registration using free-form deformations: application to breast MR images (1999)Christian
Paper 7Automatic anatomical brain MRI segmentation combining label propagation and decision fusion (2006)Sarah
Paper 15Attention-based Deep Multiple Instance Learning (2018)Indu
26.11.2021Paper 8Voxelmorph: a learning framework for deformable medical image registration (2019)Sarah
Paper 9MR Image Reconstruction From Highly Undersampled k-Space Data by Dictionary Learning (2011)Nikolas
3.12.2021Paper 10A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstruction (2018)Paul
Paper 11Generative Adversarial Networks for Noise Reduction in Low-Dose CT (2017)Stefano
10.12.2021Paper 12Learning a variational network for reconstruction of accelerated MRI data (2017)Nikolas
Paper 13Clinically applicable deep learning for diagnosis and referral in retinal disease (2018)Stefano
17.12.2021Paper 14High-throughput adaptive sampling for whole-slide image analysis (HASHI) via convolutional neural networks: Application to invasive breast cancer detection (2018)Paul
Recap session
Holidays
14.1.2022Paper 16DeepDose: Towards a fast dose calculation engine for radiation therapy using deep learning (2020)Christian
Paper 17Efficient Multi-Scale 3D CNN with Fully Connected CRF for Accurate Brain Lesion Segmentation (2016)Sarah
21.1.2022Paper 18A probabilistic u-net for segmentation of ambiguous images (2018)Nikolas
Paper 19Towards safe deep learning: accurately quantifying biomarker uncertainty in neural network predictions (2018)Paul
28.1.2022Paper 20f-AnoGAN: Fast unsupervised anomaly detection with generative adversarial networks (2019)Indu
Paper 21[cancelled] Statistical parametric maps in functional imaging: a general linear approach 1995)Paul
4.2.2022Paper 22A bi-ventricular cardiac atlas built from 1000+ high resolution MR images of healthy subjects and an analysis of shape and motion (2015)Lisa
Paper 23A multi-modal parcellation of human cerebral cortex (2016)Christian
11.2.2022Paper 24Causality matters in medical imaging (2020)Indu
Retrospective