Module CS4353-KP07

Duration

1 Semester

Turnus of offer

beginning each winter semester

Credit points

7

Course of studies, specific fields and terms:

• Master Auditory Technology 2022, compulsory module depending on previous knowledge , compulsory module depending on previous knowledge

Classes and lectures:

• CS4353-Ü: Medical Data Science (exercise, 4 SWS)
• CS4353-V: Medical Data Science (lecture, 2 SWS)

Workload:

• 50 hours exam preparation
• 70 hours private studies
• 90 hours in-classroom work

Contents of teaching:

• Introduction to Medical Data Science for Assistive Health Technologies
• General Approach to Human Activity Recognition
• Multiple Sensor Integration and Synchronisation
• Feature Learning from Multimodal Sensor Data
• Supervised Classification of Multimodal Sensor Data
• General Approach to Indoor Localisation
• Statistical Representation of Multimodal Sensor Data
• Recursive Probability Density Estimation
• Particle Filtering and State Classification
• General Approach to Sleep Lab Data Analysis
• Multimodal Time Series Data Augmentation
• Transfer Learning for Time Series Classification
• Explainable Machine Learning
• Demonstrators from Current Research Projects
• Summary and Conclusions

Qualification-goals/Competencies:

• Students have an overview of known assistive health technologies and are able to motivate their application from the medical perspective.
• Students know the general approach to human activity recognition.
• Students know selected approaches of multiple sensor integration and synchronisation.
• Students know selected feature learning methods and are able to implement them in a programming language.
• Students know selected classification algorithms for multimodal sensor data are able to implement them in a programming language.
• Students know the general approach to indoor localisation.
• Students know selected models for statistical representation of multimodal sensor data and are able to implement them in a programming language.
• Students know the theory behind the recursive probability density estimation.
• Students know the particle filtering approach and are able to implement it in a programming language.
• Students know the general approach aiming at the interpretation of data recorded in a sleep lab.
• Students know selected methods for multimodal time series data augmentation and are able to implement them in a programming language.
• Students know selected methods of explainable machine learning.
• Students know the objectives and function of software systems from selected current medical data science research projects.
• Students know the societal relevance of assistive health technologies.

Grading through:

• exercises, project, oral or written exam

Responsible for this module:

• Prof. Dr.-Ing. Marcin Grzegorzek

Teacher:

• Institute of Medical Informatics
• Prof. Dr.-Ing. Marcin Grzegorzek

Literature:

• Peter J. Brockwell and Richard A. Davis : Introduction to Time Series and Forecasting ISBN: 978-3-319-29852-8
• Marcin Grzegorzek : Sensor Data Understanding ISBN: 978-3-8325-4633-5
• Andrew R. Webb : Statistical Pattern Recognition ISBN: 978-0-470-68228-9
• Sergios Theodoridis and Konstantinos Koutroumbas : Pattern Recognition ISBN: 978-1-597-49272-0

Language:

• German or English

Notes:

Last Updated:

30.09.2025