Intelligent Biomedical Sensing 2: Deep LearningModule description #41308 / Version 1
- 9 CP
- Portfolio examination
- Graded
- Summer semester
- English
| Module / Version | #41308 / #1 |
|---|---|
| Validity | Since Summer semester 2026 |
| Available language(s) | English, German |
| Credit points | 9 |
| Person responsible for the module | von Lühmann, Alexander |
| Grading | Graded |
| Exam format | Portfolio examination |
| Language(s) of instruction | English |
Affiliation
| Faculty | Fakultät IV |
|---|---|
| Institute | Institut für Softwaretechnik und Theoretische Informatik |
| Academic chair | 34352003 - Nachwuchsgruppe Alexander von Lühmann |
Contact
| Office | MAR 4-1 |
|---|---|
| Contact person | Not specified |
| E-mail address | vonluehmann@tu-berlin.de |
| Website | Not specified |
Learning outcomes
• Understand biomedical data challenges: Analyze the specific characteristics of biomedical signal data, including noise, artifacts, non-stationarity, inter- and intra-subject variability, limited data regimes, and domain shift.
• Design deep-learning-ready pipelines: Build robust preprocessing pipelines for biomedical signals, including filtering, artifact mitigation, normalization, segmentation, and data augmentation tailored to deep learning models.
• Develop deep learning models for biosignals: Design, implement, and train deep learning architectures for time-series biomedical data, including CNNs, RNNs, and architectures inspired by current research best practices.
• Apply modern architectures: Evaluate and apply advanced deep learning approaches such as Transformers, Graph Neural Networks, and hybrid architectures for complex biomedical signal modeling tasks.
• Leverage advanced learning paradigms: Understand and use multimodal fusion, self-supervised learning, representation learning, and explainable AI techniques in biomedical contexts.
• Conduct rigorous experimental evaluation: Design proper training, validation, and testing protocols; select appropriate metrics; and critically interpret quantitative and qualitative results.
• Translate theory into practice: Implement end-to-end deep learning solutions on real biomedical datasets, from task definition to model interpretation and reporting.
• Design deep-learning-ready pipelines: Build robust preprocessing pipelines for biomedical signals, including filtering, artifact mitigation, normalization, segmentation, and data augmentation tailored to deep learning models.
• Develop deep learning models for biosignals: Design, implement, and train deep learning architectures for time-series biomedical data, including CNNs, RNNs, and architectures inspired by current research best practices.
• Apply modern architectures: Evaluate and apply advanced deep learning approaches such as Transformers, Graph Neural Networks, and hybrid architectures for complex biomedical signal modeling tasks.
• Leverage advanced learning paradigms: Understand and use multimodal fusion, self-supervised learning, representation learning, and explainable AI techniques in biomedical contexts.
• Conduct rigorous experimental evaluation: Design proper training, validation, and testing protocols; select appropriate metrics; and critically interpret quantitative and qualitative results.
• Translate theory into practice: Implement end-to-end deep learning solutions on real biomedical datasets, from task definition to model interpretation and reporting.
Content
Deep Learning for Biosignal Analysis Project
• Definition of a biomedical signal learning task (classification, regression, or segmentation)
• Literature-driven understanding of the chosen modality and task
• Design and implementation of preprocessing pipelines for deep learning
• Selection, justification, and implementation of suitable deep learning architectures
• Model training, optimization, evaluation, and systematic result analysis
• Critical reflection on limitations, generalization, and interpretability
Deep Learning for Biomedical Signal Analysis
• Introduction to deep learning in biomedical signal analysis
• Biomedical signal preprocessing and augmentation for deep learning
• Core architectures for time-series modeling (CNNs, RNNs)
• Modern architectures: Transformers, Graph Neural Networks, and beyond
• Multimodal representation learning
• Evaluation strategies, interpretability, and clinical trust
• Real-world datasets, open challenges, and current research trends
• Definition of a biomedical signal learning task (classification, regression, or segmentation)
• Literature-driven understanding of the chosen modality and task
• Design and implementation of preprocessing pipelines for deep learning
• Selection, justification, and implementation of suitable deep learning architectures
• Model training, optimization, evaluation, and systematic result analysis
• Critical reflection on limitations, generalization, and interpretability
Deep Learning for Biomedical Signal Analysis
• Introduction to deep learning in biomedical signal analysis
• Biomedical signal preprocessing and augmentation for deep learning
• Core architectures for time-series modeling (CNNs, RNNs)
• Modern architectures: Transformers, Graph Neural Networks, and beyond
• Multimodal representation learning
• Evaluation strategies, interpretability, and clinical trust
• Real-world datasets, open challenges, and current research trends
Module components
Compulsory area
The following courses are mandatory for this module:
Workload and credit points
Deep Learning for Biomedical Signal Analysis (Project)
| Workload description | Multiplier | Hours | Total |
|---|---|---|---|
| Lecture Attendance | 15 | 4 h | 60 h |
| Meetings with Supervisor(s) | 15 | 1 h | 15 h |
| Project Implementation and Project Report | 1 | 70 h | 70 h |
| Literature Research and Processing | 1 | 65 h | 65 h |
| 210 h | |||
Deep Learning for Biomedical Signal Analysis (Lecture)
| Workload description | Multiplier | Hours | Total |
|---|---|---|---|
| Attendance | 15 | 2 h | 30 h |
| Pre/post processing | 15 | 2 h | 30 h |
| 60 h | |||
The workload of the module adds up to 270 hours. The module therefore comprises 9 credit points.
Description of teaching and learning methods
Project-based learning: The graded project is the central element and focuses on applied deep learning with real biomedical data.
Lectures: Concept-driven lectures provide theoretical foundations, methodological context, and links to current research.
Hands-on coding sessions: Practical implementation using Python and PyTorch, including guided demos and exercises.
Self-study: Independent reading, implementation, experimentation, and report preparation.
Lectures: Concept-driven lectures provide theoretical foundations, methodological context, and links to current research.
Hands-on coding sessions: Practical implementation using Python and PyTorch, including guided demos and exercises.
Self-study: Independent reading, implementation, experimentation, and report preparation.
Requirements for participation and examination
Desirable prerequisites for participation in the courses:
• Solid knowledge of linear algebra, calculus, probability theory, and statistics
• Basic foundations of machine learning
• Fundamental programming skills (e. g., C/C++, Python)
• Basic foundations of machine learning
• Fundamental programming skills (e. g., C/C++, Python)
Mandatory requirements for the module examination registration:
This module has no examination requirements.
Completion of the module
| Grading | Graded |
|---|---|
| Exam format | Portfolio examination |
| Type of portfolio examination | 100 points in total |
| Language | English |
Test elements
| Name | Points/Weight | Category | Duration/Extent |
|---|---|---|---|
| (Deliverable assessment) Project report | 50 | written | 10-15 pages |
| (Deliverable assessment) Project presentation | 30 | oral | 20 minutes |
| (Examination) Multiple choice test | 20 | written | 45 minutes |
Duration of the module
The following number of semesters is estimated for taking and completing the module: 1 semester(s).
This module can be started in the following semesters: Summer semester.
Maximum number of participants
The maximum capacity of participants is 20.
Registration procedures
None.
Recommended reading, lecture notes
Lecture notes
unavailable
Electronic lecture notes
unavailable
Recommended literature
No recommended literature given
Assigned programs
This module version is used in the following programs:
| Program / program version | Semester(s) | Curriculum component |
|---|---|---|
| Computer Engineering Computer Engineering (M. Sc.) - StuPO 2015 | SS 2026 | Wahlpflicht Studiengebiete Fak. IV › Kognitive Systeme / Cognitive Systems |
| WS 2026/27 | Wahlpflichtbereich › Studiengebiet B › Kognitive Systeme / Cognitive Systems | |
| Computer Science (Informatik) Computer Science (Informatik) (M. Sc.) - StuPO 2015 | SS 2026 | Wahlpflicht Studiengebiete Fak. IV › Kognitive Systeme / Cognitive Systems |
| WS 2026/27 | Wahlpflichtbereich › Studiengebiete › Data and Software Engineering / Data and Software Engineering | |
| Electrical Engineering Electrical Engineering (M. Sc.) - StuPO 2015 | SS 2026 | Wahlpflicht Studiengebiete Fak. IV › Kognitive Systeme / Cognitive Systems |
| WS 2026/27 | Wahlpflichtbereich › Studiengebiete Fak. IV › Data and Software Engineering / Data and Software Engineering | |
| Industrial Engineering and Management Industrial Engineering and Management (M. Sc.) - StuPO 2026 | WS 2026/27 | Engineering Sciences / Specialization › AI Engineering › Data and Software Engineering |
| Information Systems Management (Wirtschaftsinformatik) Information Systems Management (Wirtschaftsinformatik) (M. Sc.) - StuPO 2017 | SS 2026 | Wahlpflichtbereich › Fachstudium Informatik › Data and Software Engineering |
| WS 2026/27 | Wahlpflichtbereich › Fachstudium Informatik › Data and Software Engineering |
Miscellaneous
None.