Computational imaging, modelling and AI in biomedicine
Postgraduate course
- ECTS credits
- 10
- Teaching semesters Spring
- Course code
- BMED365
- Number of semesters
- 1
- Teaching language
- English
- Resources
- Schedule
Course description
Objectives and Content
The objective and content of the course address:The computational mindset, imaging, modeling, machine learning, and AI in future biomedicine - ethical and regulatory aspects of AI. The course is a guided "journey" with a hands-on component through selected computational modeling techniques within biomedical and medical applications. Examples, demonstrations, and tasks will be related to in vivo imaging (MRI) and segmentation, imaging mass cytometry (IMC), biomarkers and prediction, network analysis ("patient similarity networks"), multimodal data, as well as large language models ("foundation models") within medicine and biology. Throughout the course, students will use principles and modern tools for data analysis, machine learning, and generative AI (e.g. ChatGPT) within biomedical applications. This will give the students an introduction to Python and Jupyter notebooks, use of the "cloud" for access to open data, calculations, and knowledge, as well as insight into and rationale for "open science" and "reproducible research". All course material will be openly available on a Git Hub repository: github.com/arvidl/BMED365-2026
Use of Artificial Intelligence (AI)
The use of AI tools in academic work must be guided by the principles of honesty, transparency, and academic integrity. Students are expected to follow the Faculty of Medicine’s classification system for AI use (Level G), and to clearly document their level of AI involvement in their thesis work.
Student Responsibilities
- Self-assessment: Students must assess and declare the level of AI use in their thesis, based on the official classification system.
- Critical evaluation: AI-generated content must not be used uncritically. All information must be verified, fact-checked, and integrated with academic rigor.
- Tool selection: Students are encouraged to use secure and institutionally supported AI tools. At the University of Bergen (UiB), Microsoft Copilot or UiBChat are available and considered a secure option for academic use.
For more information about the classification system and guidelines, see:
www.uib.no/en/med/174309/use-artificial-intelligence-ai-faculty-medicine
Learning Outcomes
Upon completion of the course, the student must have the following learning outcomes defined in terms of knowledge, skills, and general competence:
Knowledge
The student ...Has broad knowledge of the terms "multiscale" and "multiparametric biomedical imaging",
"computational modeling", "big data", "network analysis", "machine learning", "deep learning", and "generative AI" (incl. large language models) and was able to relate these terms to examples from biomedicine and personalized and precision medicine.
Skills
The student ...Can find and use a selection of modern software tools for data analysis, visualization, reporting, and generative AI, e.g. explorative data analysis, figure and graphics production with Jupyter notebooks, and efficient prompting of large language models such as ChatGPT. Can communicate about selected methods, software packages, and libraries where these methods have been implemented and explain their relevance in biomedical research and medicine.
General competence
The student ...Recognizes the importance of mathematical models and computational approaches, as well as large language models, for the analysis and understanding of complex systems and disease processes. Appreciate the need for interdisciplinary collaboration in biomedicine of the future. Ethical and regulatory aspects of biomedical AI. Can analyze how scientific collaboration in the form of "open science", sharing of data, and "reproducible research" can move science forward.
ECTS Credits
Level of Study
Semester of Instruction
Required Previous Knowledge
Recommended Previous Knowledge
Credit Reduction due to Course Overlap
Access to the Course
Students admitted to a Master's program at the Faculty of Medicine or the Faculty of Science and Technology at UiB (or another university) and students admitted to the engineering studies at HVL (or another university, e.g. Erasmus student). Qualified students from outside UiB will receive guest student status upon admission to the course.
There is room for max 20 students, with 10 places reserved for master's students in biomedicine.
Teaching and learning methods
The teaching style is oriented towards "blended learning" and "flipped classroom":
The course is divided into two blocks. In the first block there will be two days of introductory and motivational lectures, including demonstrations.There wil be e-learning/lab modules on a GitHub repository (before, during, and available after the course) addressing the learning outcomes for the course. A submission related to a specific topic within biomedicine, chosen from among a small selection of pre-defined projects, is mandatory. This will be organized as projects in small teams, where collaboration between students majoring in different topics in their Bachelor e.g. biology and informatics, respectively ("tandem"), is sought. This interdisciplinary team project must be presented orally at one of the last meetings in the first block. The second block will focus on computational biomedical imaging (MRI, IMC, ...) and modelling. Between the first and second block, the students will work on their own personal project, defined by the student henself within the scope of the course. This project must be submitted before the end of block two and will be presented to the rest of the students as a "speed posters". There will be a total of six "labs"/"meet-ups" with teachers and teaching assistents. Final digital exam. The course will assume the students have their own laptop (or borrow one).
Compulsory Assignments and Attendance
One compulsory team project participation, submission, and group presentation, followed by an individual project with a "speed-poster" presentation, partly with a peer assessment component. Compulsory activities are registered by the course supervisor and must be passed before the final exam.
Forms of Assessment
Team project and group presentation in block one, and submission of a personal project and presentation of corresponding "speed poster" in block two must be approved.
Final digital home exam (2 hours) with quiz and MCQ with a pass / no-pass.
Grading Scale
Assessment Semester
Reading List
The literature list and study material (on GitHub) will be ready by 01.07 for the autumn semester and 01.12 for the spring semester.
Course Evaluation
Examination Support Material
Simple, bilingual dictionary, that must be reviewable, meaning that one of the languages must be English, or a Scandinavian language.
The exam permits the use of Artificial Intelligence (AI) tools to support understanding and explanation of medical AI.
All resources are allowed, but you must specify exactly which resources you have had access to and how you used these sources (e.g., lookups, searches, prompts) in your answers.