Artificial Intelligence for Healthcare Data - Summer School
Academic Study Board of the Faculty of Engineering
Teaching language: English
EKA: T930017102
Censorship: Second examiner: Internal
Grading: 7-point grading scale
Offered in: Odense
Offered in: Summer school (spring)
Level: Bachelor
Course ID: T930017101
ECTS value: 5
Date of Approval: 07-01-2022
Duration: Intensive course
Version: Archive
Course ID
Course Title
ECTS value
5
Internal Course Code
Responsible study board
Administrative Unit
Date of Approval
Course Responsible
| Name | Department | |
|---|---|---|
| Mette Lind Johansen | melj@tek.sdu.dk | TEK Uddannelseskoordinering og Support |
| Mikkel Baun Kjærgaard | mbkj@mmmi.sdu.dk | SDU Software Engineering |
| Sofie Birch | sbirch@tek.sdu.dk | TEK Uddannelseskoordinering og Support |
Teachers
| Name | Department | City | |
|---|---|---|---|
| Jan-Matthias Braun | j-mb@mmmi.sdu.dk | SDU Applied AI and Data Science | |
| Jürgen Herp | herp@mmmi.sdu.dk | SDU Applied AI and Data Science | |
| Manuella Lech Cantuaria | mlca@mmmi.sdu.dk | SDU Applied AI and Data Science |
Programme Secretary
Offered in
Level
Offered in
Duration
Mandatory prerequisites
Completion of minimum 2 years of a related engineering programme at university level (equivalent to 120 ECTS), this also includes Computer Science and Data Science.
Recommended prerequisites
The students should have basic programming experience and have passed an introductory course in statistics and mathematics (or equivalent).
Learning objectives - Knowledge
I. Describe what role AI can have in medical applications.
II. Explain what medical data can be comprised of, how it can be organized, and why data needs to be treated differently (including storage of sensitive data and rules around the GDPR).
III. Identify which AI tools can be used and describe how the underlying methodology works.
Learning objectives - Skills
Interpret clinical data/health databases.
II. Implement the learned methodology for data preprocessing and AI tools.
III. Present results in a meaningful and concise manner.
Learning objectives - Competences
I. Analyze a problem in the health domain, select, and apply appropriate health data preprocessing, methodology and AI tools to address the problem.
II. Collect and interpret algorithmic output and present a conclusion.
Content
The course will be divided in 4 modules that will bring together technical, modeling, and contextual skills to solve real-life problems:
Module 1 - Introduction to health data
- Course introduction (AI and Machine Learning Applied in Health)
- How is data collected and stored?
- GDPR and ethical considerations
- Data cleaning, data preprocessing, data visualization
Module 2 - Using AI for disease diagnosis and early detection
- Hands-on practice: “Using medical history data to predict prostate cancer metastasis”
Module 3 – Using AI for patient monitoring based on EEG recordings
- Hands-on practice: “Seizure detection for patients with generalized epilepsy”
Module 4 – Using AI for video feed based medical device development
- Hands-on practice: “Use of wireless capsule endoscopy for detecting colorectal cancer”Exercises are performed using standard tools in the corresponding fields, I.e., in R and Python.
URL for Skemaplan
Number of lessons
Teaching Method
Combination of lectures, conceptual and programming exercises, and project work. Some of the activities will be conducted in groups.
This course grants 5 ECTS. These are divided between lectures, hands-on classes, assignments’ preparation and exam.
The following time distribution is estimated:
Lectures: 30 hours
Hands-on classes and exercises: 26 hours
Preparation for the lectures: 40 hours
Project’s preparation and presentation: 44 hours
Time of class - Two weeks in august
Teaching language
Examination regulations
Exam regulations
Name
Exam regulations
Examination is held
At the end of the summerschool course
Tests
Exam
EKA
T930017102
Name
Exam
Description
The exam is based on an overall assessment of:
- written report
- attendance (at least 80 %)
Form of examination
Project report
Censorship
Second examiner: Internal
Grading
7-point grading scale
Identification
Student Identification Card - Date of birth
Language
English
ECTS value
5
Additional information
Enrolment is limited to 25 students. If more applicants than places, applicants who meet the mandatory requirements are prioritised according to the below selection criteria:
1. Undergraduate and graduate students from partner universities (exchange); international undergraduate and graduate guest students (fee-paying); undergraduate and graduate students from other Danish universities.
2. Ph.D students from partner universities and other international Ph.D. students; other applicants.
Students are prioritised on a first-come-first-served basis, i.e. according to the time we receive your complete application.
In case a course is filled up, we try to offer you an alternative course from your list of priorities. All final decisions about admission will be sent out continually.