Meeting Banner
Abstract #1580

q-Space Deep Learning for Alzheimer’s Disease Diagnosis: Global Prediction and Weakly-Supervised Localization

Vladimir Golkov1, Phillip Swazinna1, Marcel M. Schmitt1, Qadeer A. Khan1, Chantal M.W. Tax2, Marat Serahlazau1, Francesco Pasa1,3, Franz Pfeiffer3, Geert Jan Biessels4, Alexander Leemans5, and Daniel Cremers1

1Department of Informatics, Technical University of Munich, Munich, Germany, 2CUBRIC, Cardiff University, Cardiff, United Kingdom, 3Physics Department, Technical University of Munich, Munich, Germany, 4Department of Neurology, University Medical Center Utrecht, Utrecht, Netherlands, 5Image Sciences Institute, University Medical Center Utrecht, Utrecht, Netherlands

Most diffusion MRI approaches rely on comparably long scan time and a suboptimal processing pipeline with handcrafted physical/mathematical representations. They can be outperformed by recent handcrafted-representation-free methods. For instance, q-space deep learning (q-DL) allows unprecedentedly short scan times and optimized voxel-wise tissue characterization. We reformulate q-DL such that it estimates global (i.e. scan-wise rather than voxel-wise) information. We use this formulation to distinguish Alzheimer’s disease (AD) patients from healthy controls based solely on raw q-space data without handcrafted representations such as DTI. Classification quality is very promising. Weakly-supervised localization techniques indicate that the neural network attends to AD-relevant brain areas.

This abstract and the presentation materials are available to members only; a login is required.

Join Here