Understanding 3D CNN Behavior for Alzheimer's Disease Diagnosis from Brain PET Scan
This work addresses the interpretability problem for clinicians using CNN-based automated diagnosis systems, though it appears incremental in applying existing visualization methods to a specific medical domain.
The authors tackled the lack of transparency in 3D CNNs for Alzheimer's disease diagnosis from brain PET scans by developing a network and applying five visualization techniques to highlight relevant areas, but no concrete performance numbers were provided.
In recent days, Convolutional Neural Networks (CNN) have demonstrated impressive performance in medical image analysis. However, there is a lack of clear understanding of why and how the Convolutional Neural Network performs so well for image analysis task. How CNN analyzes an image and discriminates among samples of different classes are usually considered as non-transparent. As a result, it becomes difficult to apply CNN based approaches in clinical procedures and automated disease diagnosis systems. In this paper, we consider this issue and work on visualizing and understanding the decision of Convolutional Neural Network for Alzheimer's Disease (AD) Diagnosis. We develop a 3D deep convolutional neural network for AD diagnosis using brain PET scans and propose using five visualizations techniques - Sensitivity Analysis (Backpropagation), Guided Backpropagation, Occlusion, Brain Area Occlusion, and Layer-wise Relevance Propagation (LRP) to understand the decision of the CNN by highlighting the relevant areas in the PET data.