Visualizing Convolutional Networks for MRI-based Diagnosis of Alzheimer's Disease

TL;DR

This study trains a 3D CNN to detect Alzheimer's from MRI scans and compares four visualization methods to interpret the model's focus areas, enhancing trust in AI-based medical diagnosis.

Contribution

It introduces a comprehensive comparison of gradient-based and occlusion-based visualization methods for CNN interpretability in Alzheimer's diagnosis.

Findings

All methods focus on known Alzheimer's brain regions.

Gradient methods reveal distributed relevance patterns.

Relevance distribution varies across patients.

Abstract

Visualizing and interpreting convolutional neural networks (CNNs) is an important task to increase trust in automatic medical decision making systems. In this study, we train a 3D CNN to detect Alzheimer's disease based on structural MRI scans of the brain. Then, we apply four different gradient-based and occlusion-based visualization methods that explain the network's classification decisions by highlighting relevant areas in the input image. We compare the methods qualitatively and quantitatively. We find that all four methods focus on brain regions known to be involved in Alzheimer's disease, such as inferior and middle temporal gyrus. While the occlusion-based methods focus more on specific regions, the gradient-based methods pick up distributed relevance patterns. Additionally, we find that the distribution of relevance varies across patients, with some having a stronger focus on the temporal lobe, whereas for others more cortical areas are relevant. In summary, we show that applying different visualization methods is important to understand the decisions of a CNN, a step that is crucial to increase clinical impact and trust in computer-based decision support systems.