Module CS4371 T

Duration

1 Semester

Turnus of offer

each winter semester

Credit points

8

Course of studies, specific fields and terms:

• Master MES 2020, module part, computer science / electrical engineering
• Master MES 2014, module part, computer science / electrical engineering

Classes and lectures:

• Fortgeschrittene Verfahren der Med. Bildverarbeitung (practical course, 1 SWS)
• Fortgeschrittene Verfahren der Med. Bildverarbeitung (lecture, 3 SWS)
• Fortgeschrittene Verfahren der Med. Bildverarbeitung (exercise, 2 SWS)

Workload:

• 60 hours private studies and exercises
• 30 hours exam preparation
• 60 hours private studies
• 90 hours in-classroom work

Contents of teaching:

• Applications of medical image processing techniques
• Image superresolution
• Denoising and inhomogeneity correction
• Linear and non-linear dimensionality reduction
• Patch-based image processing and non-local means
• Fusion of (probabilistic) segmentations (NLM and STAPLE)
• Random-walk algorithm for interactive segmentation
• Non-linear registration and motion estimation (optical flow)
• Similarity metrics for multi-modal fusion
• Introduction into graphical models and discrete optimisation
• Viterbi algorithm and message passing (stereo depth estimation)
• Graph cut segmentation and further applications
• Extraction image features and descriptors
• Matching of corresponding landmarks

Qualification-goals/Competencies:

• Students know a wide range of methods for segmentation, registration and processing of medical images.
• They can describe these methods with correct technical terminology.
• They can transfer image processing techniques into energy minimisation problems.
• They can solve minimisation problems using sparse linear systems.
• They understand methodological relations between different applications and techniques.
• They understand the transfer of continuous problems into the discrete domain.
• They understand solvers for discrete optimisation problems.
• They can transfer mathematical concepts into practical algorithms for medical image processing.
• They can proficiently implement these concepts in C++.
• They can compare different algorithms to another and make suitable problem-related choices of methods.
• They have an extended overview of application areas for medical image analysis.

Grading through:

• Oral examination

Requires:

• Module part: Medical Image Computing (CS3310 T)

Responsible for this module:

• Siehe Hauptmodul

Teacher:

• Institute of Medical Informatics
• Prof. Dr. Mattias Heinrich

Literature:

• M. Sonka, V. Hlavac, R. Boyle : Image Processing, Analysis and Machine Vision 2nd edition. Pacific Grove: PWS Publishing 1998

Language:

• offered only in German

Notes:

Last Updated:

24.09.2021