Zurück zur Auswahl
53822-01 - Vorlesung: Deep Learning for Medical Image Analysis (3 KP)
| Semester | Frühjahrsemester 2026 |
| Angebotsmuster | Jedes Frühjahrsem. |
| Dozierende |
Florentin Bieder (florentin.bieder@unibas.ch)
Philippe Claude Cattin (philippe.cattin@unibas.ch, BeurteilerIn) |
| Inhalt | This course provides an introduction to deep learning and how this cutting-edge technology can be applied to medical image analysis with exercises using PyTorch. The course covers the following topics * gradient based optimization * backpropagation * multilayer perceptrons (MLPs) * convolutions and convolutional Neural Networks (CNNs) * network building blocks (activations, pooling, normalization, etc.) * losses and metrics for regression and classification problems * parameter initialization heuristics * data preprocessing and encoding * data augmentation * resource consumption * common architectures (transformers, generative adversarial models, diffusion models, etc.) |
| Lernziele | * understand the basics of deep learning and how it can be applied to medical image analysis * understand numerical optimization algorithms used to train deep learning models * understand the architecture and training of multilayer perceptrons and CNNs * medical applications of MLPs and CNNs for classification, regression, segmentation, and anomaly detection tasks * know different generative models and their medical applications |
| Weblink | DBE MA BME |
| Teilnahmevoraussetzungen | (C15) Medical Imaging and Medical Image Processing; Python Knowledge similar to course 69472 Limited student numbers, priority given to student in Biomedical Engineering |
| Unterrichtssprache | Englisch |
| Einsatz digitaler Medien | kein spezifischer Einsatz |
| Intervall | Wochentag | Zeit | Raum |
|---|---|---|---|
| wöchentlich | Mittwoch | 15.15-17.00 | Hegenheimermattweg 167B, Lecture Hall 02. 097 |
Einzeltermine
| Module |
Modul: Biomedical Engineering Electives (Masterstudium: Biomedical Engineering) Modul: Electives in Data Science (Masterstudium: Data Science) Modul: Vertiefung Medizinische Nanowissenschaften (Masterstudium: Nanowissenschaften) |
| Prüfung | Leistungsnachweis |
| Hinweise zur Prüfung | written exam |
| An-/Abmeldung zur Prüfung | Anm.: Belegen Lehrveranstaltung; Abm.: stornieren |
| Wiederholungsprüfung | eine Wiederholung, bester Versuch zählt |
| Skala | 1-6 0,1 |
| Belegen bei Nichtbestehen | beliebig wiederholbar |
| Zuständige Fakultät | Medizinische Fakultät |
| Anbietende Organisationseinheit | Departement Biomedical Engineering (DBE) |