The course covers a number of machine learning methods and concepts, including state-of-the-art deep learning methods, with example applications in the medical imaging and computational biology domains.
The lectures are mainly based on the selected chapters from following two books that are freely available online:
- Deep Learning, Ian Goodfellow and Yoshua Bengio and Aaron Courville
- Elements of Statistical Learning, Trevor Hastie, Robert Tibshirani, Jerome Friedman
Additional reading materials such as journal articles are listed within the lecture slides.
- [Lecture slides]
- [Practical work]
- [Lecture slides]
- [Practical work]
- [Lecture slides]
- [Practical work]
- [Lecture slides]
- [Practical work]
- [Lecture slides]
- [Practical work]
- [Reading list]
After completing the course, the student will be able to:
- Recognize how machine learning methods can be used to solve problems in medical imaging and computational biology.
- Comprehend the basic principles of machine learning.
- Implement and use machine learning methods.
- Design experimental setups for training and evaluation of machine learning models.
- Analyze and critically evaluate the results of experiments with machine learning models.
The assessment will be performed in the following way:
- Work on the practical assignments: 25% of the final grade (each assignment has equal contribution);
- Unuspervised learning presentations: 10% of the final grade;
- Final exam: 65% of the final grade.
Intermediate feedback will be provided as grades to the submitted assignments.
The grading of the assignments will be done per groups, however, it is possible that individual students get separate grade from the rest of the group (e.g. if they did not sufficiently participate in the work of the group).
The students will receive instruction in the following ways:
- Lectures;
- Guided practical sessions;
- Contact hours with the project instructors for questions, assistance and advice;
- Online discussion (in Canvas, see below).
Course instructors:
- Mitko Veta
- Federica Eduati
8DB00 Image acquisition and Processing, and 8DC00 Medical Image Analysis.
The practical assignments for this course will be done in Python. We recommend the Anaconda Python distribution.
The [course page in Canvas] will be used for submission of the assignments, scheduling of the lectures and contact hours and announcements. The students are highly encouraged to use the Discussion section in Canvas. All general questions (e.g. issues with setting up the programming environment, error messages etc., general methodology questions) should be posted in the Discussion section in Canvas and not asked during the contact hours.
This page is carefully filled with all necessary information about the course. When unexpected differences occur between this page and Osiris, the information provided in Osiris is leading.