| Lecture & Tutorials: Digital Image Processing I - Information carrier image: Introduction to methodology, technology and applications of digital image processing
- Imaging: the human eye, cameras, other sensors and imaging geometry
- Image pre-processing: grids and interpolation methods, homogeneous and inhomogeneous point operations
- Filtering: 2D Fourier transform, FIR filters and non-linear filters, feature extraction: edge detection, gradient and Laplace filters
- Segmentation: point-based, region-based, contour-based, model-based and using neural networks
- Registration: point registration, surface registration, elastic registration
- Visualisation of 3D image data: indirect and direct volume rendering
Lecture & Tutorials: Digital Image Processing II - Advanced U-Net Architectures
- Attention Mechanisms and Transformers
- Hybrid U-Net / Transformer Architectures
- Generative Models (e.g. Auto-Regressive Models, Variational Autoencoders, Generative Adversarial Networks, Energy-based Models, Normalizing Flows, Diffusion models)
- Semi- and Unsupervised Learning Techniques
- Foundation Models
- Continual Learning
- Emerging Trends and Clinical Applications
Self-Paced Research - In this module, you will learn to conduct independent scientific research in data science. Working in teams of 3-4, you will tackle a research question over the course of the semester, developing and adjusting project plans as you go. You will meet regularly with a mentor for feedback and present your progress twice per semester to your peers and mentors. Through hands-on projects focused on current data science topics, you will gain skills in project planning, documentation, and scientific communication.
Seminar - Advanced Deep Learning Strategies for Medical Image Analysis
|
