Hasso-Plattner-Institut
HPI Digital Health Center
  
 

Applied Machine Learning for Digital Health

General Information

News

Topics and selection procedure will be presented during the kickoff event.

Slides

Lecture

Scope of the lecture

This seminar helps you to broaden your expertise in Machine Learning (ML). Therefore, well will introduce to you selected ML technologies and tools for your topics. You will acquire hands-on experience with these tools and apply them to real-world scenarios. We will equip you with the required machine learning expertise and provide you access to materials for your selected topics. We expect from you to deep dive in the required ML technology, to learn about related work, to design and apply your own ML approach, and to evaluate your approach and compare it to results from related work.   

You will select your preferred topic for a list of topics presented in the kick-off lecture. We will coach you throughout the semester and will help to improve your research and presentation skills. You will share your results in an intermediate and final presentation together with all participants. You will document your final results in a scientific way.

Grading

The final grading will be determined by the following individual parts, while each part must be passed individually: 

  • Intermediate presentation, final presentation (40%)
  • Research article (40%)
  • Individual commitment (20%)

Topics

Topic Who?
1. Computer Vision (Benjamin Bergner)  
A) CT organ segmentation for radiotherapy planning of lung cancer  
B) CT liver segmentation for living donor liver transplantations  
2. Human Activity Recognition (Orhan Konak)  
A) Comparison of Models  
B) Preprocessing  
3. Biomedical Natural Language Processing (Florian Borchert)  
A) Weak supervision for medical information extraction  
B) Pre-trained language models for medical information extraction  
4. Clinical Predictive Modeling (Harry Freitas da Cruz)  
A) Novel metrics for clinical predictive models  
B) Synthetic data using generative adversarial networks