Goal

Become acquainted with the implementation of image filtering in the spatial and frequency domain, the application of image restoration and color image processing algorithms, finding image attributes (points, edges, etc.) and features, controlling segmentation parameters for obtaining optimal results and gain a basic understanding of machine learning algorithms and their validation.

Contents

Medical imaging is the most important discipline for healthcare diagnosis and intervention. This course provides in-depth knowledge on medical imaging techniques and the principles of the medical image analysis. This course starts with an introduction to sampling and the extension to multi-dimensional signal processing for images and video. Then the course proceeds with the principles and definitions for the frequency domain representation of images and possibilities for filtering and other processing in the frequency domain. In a further module, the focus is on the image quality enhancement in various ways. Various noise models are discussed and filtering approaches for reducing the noise. The introduction of color in medical images for improved understanding is addressed. The second part of the course provides an introduction to image understanding and object/area segmentation to learn imaging for medical diagnosis. This part starts with simple binary operators and then extends to the detection of points, lines, edges, etc. Secondly, specific region-based segmentation methods are discussed. Features are extracted from images, which is addressed in both the spatial domain directly in the image and a frequency-oriented domain, such as the Gabor transformation. The course concludes with a module on imaging techniques and use cases, where the learnt techniques are applied. These cases include computer-aided cancer diagnosis, instrument detection, etc.

Preknowledge:

5ESA0 Signal processing basics