Course syllabus for Artificial Intelligence for Clinical and Biomedical Applications

Artificiell intelligens för kliniska och biomedicinska applikationer

Specific entry requirements

At least the grade Pass on the courses Integrated physiology and pharmacology (semester 1) and Physiological and pharmacological mechanisms and experimental methods (semester 2) on the Master programme in Translational Physiology and Pharmacology.

Outcomes

The aim of the course is to provide an overview of the applicability and use of ML/AI (Machine Learning/Artificial Intelligence) in Life Sciences with the intention to deepen the student's understanding of the use of ML/AI in image analysis, drug discovery, diagnostics and prognostics.

After completing the course, the student shall be able to:

• Describe core ML and AI concepts and their role in Life Science
• Gain an overview of ML/AI platforms and complete a guided hands-on exercise
• Discuss the implications of the use of ML/AI in drug discovery, image analysis, diagnostics and prognostics
• Have a critical and ethical approach to the use of ML/AI in Life Sciences
• Describe core concepts in machine learning, deep learning, and large language models
• Describe how AI tools like AlphaFold accelerate drug discovery and protein structure prediction
• Show practical skills in handling and analyzing clinical datasets and medical images
• Describe AI's potential role in analyzing clinical datasets from HumanLab and similar human physiology experiments
• Discuss major ethical, regulatory, and safety considerations in medical AI
• Critically evaluate AI models’ reliability, fairness, and clinical relevance

Content

The course introduces core concepts of ML and AI and their implications for image analysis, drug discovery, diagnostics, prognostics and HumanLab data analysis. The focus areas are:

• Foundations of Medical AI: overview machine learning, deep learning, and generative AI in health care.
• AI in diagnostics and prognostic models: Disease diagnosis and outcome prediction using clinical and imaging data (e.g. CNN models for radiology).
• Introduction to concepts of AI in drug discovery: Overview of AI roles in target identification, QSAR modeling, large scale* in silico* docking screens, de novo peptide design and protein folding tools like AlphaFold.
• Clinical big data: Understanding data preprocessing and responsible analysis of clinical data, handling biases and missing data.
• Humanlab data applications: How AI can be used to analyze HumanLab datasets for personalized predictions.
• Ethical and regulatory aspects: Data privacy, fairness, bias mitigation, and regulatory guidelines for clinical AI deployment pathways.

The course includes theoretical and practical aspects of the use of ML/AI in Life Sciences.

Teaching methods

Learning activities include lectures, seminars and practical exercises.

Examination

Examinations

Individual written assignment. Graded Fail/Pass/Pass with distinction.

Oral group presentation of project work. Graded Fail/Pass.

Mandatory attendance

Practical exercises

Written assignments should be submitted before the end of the course according to the specification in the schedule. To pass the course (the grade Pass or higher), at least passed on all components of the course is required. To pass the course with distinction, the grade Pass with distinction on the individual written assignment is also required.

Compulsory participation
The examiner assesses if, and how, absence from compulsory course elements can be made up for. Study results cannot be reported until the student has participated in compulsory course elements or compensated for any absence in accordance with instructions from the examiner. Absence from a compulsory course element could mean that the student can not retake the element until the next time the course is offered.

Limitation of number of tests
The students that have not passed after the regular examination session have the right to participate at further five examination sessions. If the student has carried out six failed examinations/tests, no additional examination or new course admission is approved.

Each occasion the student participates in the same test counts as an examination. Submission of a blank examination is regarded as an examination. In case a student is registered for an examination but does not attend, this is not regarded as an examination.

In the event of special circumstances, or a need of adjustments for a student with a disability, the examiner may decide to depart from the syllabus' regulations on examination form, number of examination opportunities, possibility of completion or exemption from compulsory educational elements, etc. Content and intended learning outcomes as well as the level of expected skills, knowledge and abilities must not be altered, removed or lowered.

Other directives

The course is given in English, and the examinations are in English.
A course evaluation will be carried out in accordance with the guidelines established by Karolinska Institutet.

Literature and other teaching aids

Study material and reference articles will be provided during the course.