AI for Medical Image Analysis

Instructors

1. Prof Phaneendra K. Yalavarthy, Ph.D. (yalavarthy[AT]iisc.ac.in)
2. Prasad S. Murthy, Ph.D. (psudhaka[AT]gmail.com)

Office: Room No-305 (CDS); Office Hours: 3:30PM–5:00PM (Friday)

Teaching assistant

• Mr. Naveen Paluru, Ph.D. student of CDS (naveenp[AT]iisc.ac.in)

Schedule

3:30PM - 5:00PM on Tuesdays and Thursdays; Place: Room No: 202 of CDS.

Resources¹

1. Probabilistic Machine Learning: An Introduction (Adaptive Computation and Machine Learning series), Kevin P. Murphy, The MIT Press, March 2022.
2. Deep Learning Illustrated: A Visual, Interactive Guide to Artificial Intelligence, Jon Krohn, Grant Beyleveld, and Aglae Bassens – Addison Wesley, 2019.

Reading materials are posted on DS261 MS Teams page

Course outline

Artificial Intelligence (AI) for Medical Image Analysis is a current research area, where new developments are taking place every single day! This course covers the theory and technology (tools) parts. The other part of clinical implementation will not be covered in this course. Project will involve development of a compact solution to current problem/s in medical imaging using AI, such that it will enhance your understanding of challenges related to medical imaging.

Grading

Homeworks: 15%, Journal Paper Presentations: 10%, Midterm Exam: 25%, Project: 25%, and Final Exam: 25%.

Assignments

There will be 4 homework problem sets with a specific due date and time. All homework problems require computer programming. Best three of them will be used for the grading. There will be one journal paper presentation, where the student will present the assigned paper. This presentation will be evaluated on the comprehensiveness and breadth of the material covered. Each student will either choose or come with a project proposal by the end of first six weeks from start of the course. These proposals will be evaluated based on relevance and feasibility for proceeding further. Before three weeks of completion of course (last week of November), final report of the project should be submitted. Late submission (beyond the due date and time) of homework solutions/project report will result in no credit. Final project presentations will be scheduled in the first week of December.

Midterm/Final exam

Both midterm and final exam will test primarily the theory part of the course. Students will be allowed to carry one sheet of paper (A4 size) to the exam, which can contain important formulas/constants. A model exam will be posted on the course webpage for reference at least two weeks before the exam. Note that the course material that is covered in the midterm exam will not be part of the final exam.

Honour principle

You are welcome to exchange ideas in solving homework problems with your colleagues, but all the work submitted for grading (homework, presentations, project, and midterm/final exam) must be your own work (i.e., you must have worked out all details by yourself). Copying computer code or files (including the material on the web) without proper citation is considered as plagiarism. Any deviation from this principle will result in failing of the class.