Syllabus database for doctoral courses
-
Startpage
Syllabus database for doctoral courses
SYLLABI FOR DOCTORAL COURSES
| Swedish title | Artificiell intelligens och maskininlärande för biomedicinsk och klinisk forskning |
|---|---|
| English title | Artificial Intelligence and Machine Learning for Biomedical and Clinical Research |
| Course number | 5223 |
| Credits | 3.0 |
| Responsible KI department | Institutionen för mikrobiologi, tumör- och cellbiologi |
| Specific entry requirements | At least 1,5 credits from a course in basic statistics. |
| Grading | Passed /Not passed |
| Established by | The Committee for Doctoral Education |
| Established | 2020-03-10 |
| Purpose of the course | To increase knowledge about Machine Learning (ML) and Artificial Intelligence (AI) applications in biological and medical research, introduce first-hand experience and skills with different frameworks. The course requires no preliminary programming skills as well as no preliminary expertise in ML and AI. This course is given at a basic/novice level with no expertise in ML/AI and preliminary programming skills required, though experience in data analysis using RStudio/MatLab or similar analytic environment is an advantage. |
| Intended learning outcomes | After the completed course, the participants will be able to describe and discuss general aspects of ML and AI in a biomedical or medical context including ethical dilemmas and challenges. Practically, they should be able to prepare and analyse different data types related to own research, such as texts, omics, genomic sequences, images etc. using a range of ML and AI exploration and classification techniques as well critically analyse the outcome and estimate performance. |
| Contents of the course | Basic information about AI and ML, multivariate dataset preparation, classic methods of univariate and multi-dimensional analysis (Principal Component Analysis, Linear Discrimination Analysis, Factor Analysis), variable selection and sparse regression models (lasso regression, ridge regression, elastic net), supervised and unsupervised learning with neural networks, federated learning, performance estimation methods. |
| Teaching and learning activities | The course consists of lectures, group discussions, and hands-on labs. Previous experience from practical experience applying modelling in a computer-based environment (e.g. in R, SAS, STAT, Matlab or Python), is strongly recommended. |
| Compulsory elements | All planned activities including lab and group works are mandatory. Absence has to be compensated with a report on the lab work, which student will have to do. |
| Examination | The student will be examined by their (a) labs accomplishment (b) final project report and (c) written reviews of projects of 2 other students. |
| Literature and other teaching material | Both classic and up-to-date articles and websites will be recommended. | Course responsible |
Iurii Petrov Institutionen för mikrobiologi, tumör- och cellbiologi iurii.petrov@ki.se |
| Contact person |
Matti Nikkola Institutionen för cell- och molekylärbiologi Matti.Nikkola@ki.se |
