MaUBRL

Chair for Computer Aided Medical Procedures & Augmented Reality
Lehrstuhl für Informatikanwendungen in der Medizin & Augmented Reality

Unsupervised Brain Representation Learning

Supervision: Prof. Dr. Nassir Navab, Dr. Seong Tae Kim, Matthias Keicher

Abstract

In this project, we investigate a method to learn an unsupervised representation of the brain by exploiting the transformation equivariance and interpretability properties of capsules networks. The unsupervised representations obtained by our network are evaluated by performing downstream classification and anomaly detection tasks on the latent space. In addition, we will further explore the feature disentanglement and image generation capabilities of our network.

Requirements:

  • Good understanding of statistics and machine learning methods.
  • Very good programming skills in Python & TensorFlow? / PyTorch?

Location:

  • Garching

ProjectForm
Title: Unsupervised Brain Representation Learning
Abstract: In this project, we investigate a method to learn an unsupervised representation of the brain by exploiting the transformation equivariance and interpretability properties of capsules networks. The unsupervised representations obtained by our network are evaluated by performing downstream classification and anomaly detection tasks on the latent space. In addition, we will further explore the feature disentanglement and image generation capabilities of our network.
Student:  
Director: Prof. Dr. Nassir Navab
Supervisor: Dr. Seong Tae Kim, Matthias Keicher
Type: Master Thesis
Area: Machine Learning, Medical Imaging
Status: running
Start:  
Finish:  
Thesis (optional):  
Picture:  


Edit | Attach | Refresh | Diffs | More | Revision r1.1 - 28 Apr 2020 - 19:14 - SeongTaeKim