Skip to content
Technische Universität Berlin Deutsch
Course catalog and module transfer system

Machine Learning and Inverse Problems in NeuroimagingModule description #41068 / Version 2

← Module page Deutsche Fassung
This module description is not valid in the current semester (Winter semester 2026/27). Current version (v5)
Module / Version#41068 / #2
ValiditySummer semester 2023 – Summer semester 2023
Available language(s)English, German
Credit points3
Person responsible for the moduleNot specified
GradingGraded
Exam formatPortfolio examination
Language(s) of instructionEnglish

Affiliation

FacultyFakultät IV
InstituteInstitut für Softwaretechnik und Theoretische Informatik
Academic chair34352700 FG S-Professur Unsicherheit, inverse Modellierung und maschinelles Lernen (UNIML)

Contact

OfficeMAR 4-3
Contact personHaufe, Stefan
E-mail addresshaufe@tu-berlin.de
Websitehttps://www.tu.berlin/uniml/

Learning outcomes

Knowledge: Students will understand the basic concepts of statistical and physical forward and inverse problems. They will be familiar with the physics of magneto- and encephalography (M/EEG) measurements and the steps to calculate M/EEG forward models. They will know different mathematical frameworks for inverse modeling, such as dipole fitting, beamforming, distributed imaging, blind source separation, and Bayesian modeling. They will be familiar with the concept of regularization and ways to encode prior knowledge into inverse solvers. They will learn to address technical challenges such as estimating the noise level/choosing the regularization parameter.

Skills: Students will be capable of modeling inverse problems as unsupervised or supervised machine learning problems. Students will also acquire or refine skills to independently review and systematically structure the literature of a well circumscribed field in order to address a given set of questions, and will gain experience in presenting the outcome to a critical audience as well as in participating in scientific discussions.

Competencies: Students will be able to discuss the advantages and disadvantages of different modeling approaches depending on the problem setting and to make informed modeling decisions.

Content

- common inverse problems in biomedicine, in particular neuroimaging
- physical foundations of magneto-/electroencephalography (M/EEG)
- forward modeling and physics simulation for M/EEG
- dipole fits, beamforming, scanning techniques
- penalized likelihood approach: smoothness, structured sparsity, elastic net, total variation denoising
- Bayesian inference: maximum a-posteriori estimation, hierarchical and empirical Bayes, sparse Bayesian learning
- noise learning approaches
- blind source separation as a statistical inverse problem
- simulations and validation
- research software
- applications in brain-computer interfacing and neurology

Module components

Compulsory area

The following courses are mandatory for this module:

CoursesTypeNumberFrequencyLanguageHours per weekCourse catalog
Machine Learning and Inverse Modeling in NeuroimagingSeminar Irregular[en]2 ISIS

Workload and credit points

Machine Learning and Inverse Modeling in Neuroimaging (Seminar)

Workload descriptionMultiplierHoursTotal
Attendance152 h30 h
Pre/post processing154 h60 h
90 h

The workload of the module adds up to 90 hours. The module therefore comprises 3 credit points.

Description of teaching and learning methods

The module consists of a single seminar. Students will prepare a presentation to a specific topic based on a provided collection of published material. Student presentations will be framed by short lecture segments introducing, contextualizing and connecting the presented topics. Each course slot will contain discussion periods, in which active participation is fostered. All students will also prepare brief summaries for each topic.

Requirements for participation and examination

Desirable prerequisites for participation in the courses:

A BSc degree in Computer Science, Biomedical Engineering or a comparable field is recommended.

Basis knowledge of linear algebra, stochastics, and numerical optimization is advantageous.

Mandatory requirements for the module examination registration:

This module has no examination requirements.

Completion of the module

GradingGraded
Exam formatPortfolio examination
Type of portfolio examination100 points in total
LanguageEnglish

Test elements

NamePoints/WeightCategoryDuration/Extent
(Deliverable assessment) Presentation50oral30 min
(Deliverable assessment) Topic summaries50written7-12 pages

Exam description

The module grade is calculated based on

1. The quality of a scientific presentation (50%).
2. The quality of short summaries for each topic (50%).

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: Winter semester.

Maximum number of participants

The maximum capacity of participants is 28.

Registration procedures

Students can sign up for the course in MOSES.

Recommended reading, lecture notes

Lecture notes

unavailable

Electronic lecture notes

unavailable

Recommended literature

No recommended literature given

Assigned programs

This module is not used in any program.

Miscellaneous

Not specified