| Dozent/in |
Ass.-Prof. Christian Frederik Baumgartner |
| Veranstaltungsart |
Vorlesung |
| Code |
FS241638 |
| Semester |
Frühjahrssemester 2024 |
| Durchführender Fachbereich |
Gesundheitswissenschaften |
| Studienstufe |
Master |
| Termin/e |
Mo, 26.02.2024, 08:15 - 10:00 Uhr, 3.B57
Mo, 26.02.2024, 14:15 - 16:00 Uhr, 3.B57
Mo, 04.03.2024, 08:15 - 10:00 Uhr, 3.B57
Mo, 11.03.2024, 08:15 - 10:00 Uhr, 3.B57
Mo, 18.03.2024, 08:15 - 10:00 Uhr, 3.B57
Mo, 25.03.2024, 08:15 - 10:00 Uhr, 3.B57
Mo, 08.04.2024, 08:15 - 10:00 Uhr, 3.B57
Mo, 15.04.2024, 08:15 - 10:00 Uhr, 3.B57
Mo, 22.04.2024, 08:15 - 10:00 Uhr, 3.B57
Mo, 29.04.2024, 08:15 - 10:00 Uhr, 3.B57
Mo, 06.05.2024, 08:15 - 10:00 Uhr, 3.B57
Mo, 13.05.2024, 08:15 - 10:00 Uhr, 3.B57
Mo, 27.05.2024, 08:15 - 10:00 Uhr, 3.B57 |
| Umfang |
2 Semesterwochenstunden |
| Inhalt |
Machine learning (ML) research is moving at a very rapid pace. Many of the recent developments in the automated analysis of images, text, as well as other data modalities has the potential to substantially reform healthcare. In this environment independently reading academic research papers, as well as a basic understanding of modern ML techniques are crucial.
In this seminar, students will learn to read, understand, and present research paper on the topic of ML for healthcare.
Each two hour session will see the presentation of some background by the instructor, as well as a student presentation on a applied ML for healthcare research paper, followed by a group discussion.
Throughout, the semester we will go from simple applications of regression models, to more advanced techniques based on neural networks.
Using the research papers as a red thread, we will see different medical data types (tabular data, text data, image data, time-series data), different modern ML tools (e.g. CNNs, Transformers, CLIP models) as well as different application areas of ML for health (data generation, risk stratification, decision support, workflow support, medical discovery). |
| Lernziele |
After completing this seminar the students will…
- be able to independently read, present and critically discuss academic research papers on machine learning for healthcare.
- understand the basics of modern machine learning techniques and how they can be applied to healthcare problems.
- develop an understanding of some clinical application areas where machine learning is likely to make an impact in the coming years.
- be ready to dive deeper in one of the discussed machine learning techniques for example with a hands-on research project. |
| Voraussetzungen |
Basic knowledge in probability theory and linear algebra is recommended as a prerequisite for this course. |
| Sprache |
Englisch |
| Begrenzung |
Important: The course is limited to 11 participants. The limit is administered via MOODLE according to chronological order and registration. From 5 February 2024, 00:00, it will be possible to register via MOODLE. As soon as 11 participants are enrolled, the registration window will be closed automatically. If you wish to be put on the waiting list, then please send an email to: masterhealth@unilu.ch |
| Anmeldung |
https://elearning.hsm-unilu.ch/course/view.php?id=735 |
| Leistungsnachweis |
The final grade consists of a score for the presentation (70%), the participation as discussion leader (20%), as well as the general participation throughout the course (10%). |
| Abschlussform / Credits |
The final grade consists of a score for the presentation (70%), the participation as discussion leader (20%), as well as the general participation throughout the course (10%). / 3 Credits
|
| Hinweise |
Teaching methods:
Most sessions consist of a presentation by the instructor, as well as a student presentation, followed by a joint group discussion.
Each student will present one research paper throughout the course. Moreover, each student will be assigned as discussion leader for two additional research papers. The responsibility of the discussion leader is to understand the respective papers in detail and come up with critical and engaging discussion topics.
Presence is required for at least 11 out of the 13 sessions.
The seminar is limited to 11 participants. |
| Hörer-/innen |
Nein |
| Material |
The presentation slides of the instructor as well as the individual student presentations will be made available to all students. |
| Literatur |
The research papers will be made available to the students ahead of the course. |