Contents
Classification and prediction are major aspects of health science. The course aims to introduce recent methodology of statistical learning that has proven useful at least in assisted diagnostics and prognosis, useful in detecting patterns and is expected to play an increasing role in clinical settings.
Main topics to be covered:
- Artificial intelligence - what is it?
- Design of studies using AI
- Evaluation of classification and prediction using AI
- The patient and AI
Aim
The participant will obtain knowledge and understanding of modern statistical AI learners. Competence to identify basic design, relevant analysis and to evaluate the outcome in settings of both research and practice. Demonstrations of certain AI learners, neural networks will be given and discussed. Material will be available for trying out assisted classification and prediction, but model application skills are not required. Feedback on participants approaches to AI in research or practice will be provided.
Teaching arrangement
Lectures, worked examples, exercises and web-based multiple choice exam
Duration
2 days, 9.15-15.00
Exam
Take-home lesson with questions on concepts. To be handed in before five working days from last day of the course.
Course teachers
- Ulrich Halekoh, associate professor, Institute of Public Health, SDU
- Jacob Hjelmborg, professor, Institute of Public Health, SDU
Max number of participants
36 PhD students
Course fee
The course is free of charge for PhD students enrolled in Universities that have joined the "Open market agreement".
For other participants there is a course fee of DKK 3501,-