INSightR-Net: Interpretable Neural Network for Regression using Similarity-based Comparisons to Prototypical Examples

TL;DR

This paper introduces INSightR-Net, an interpretable CNN for regression that uses similarity-based comparisons to prototypes, enabling visualization of decision reasoning in medical imaging tasks like diabetic retinopathy grading.

Contribution

It presents a novel CNN architecture that combines interpretability with competitive performance by incorporating a prototype layer for visual explanations.

Findings

Achieved comparable accuracy to ResNet baseline.

Provided meaningful visual explanations through prototype similarity.

Analyzed the impact of parameters on explanation quality.

Abstract

Convolutional neural networks (CNNs) have shown exceptional performance for a range of medical imaging tasks. However, conventional CNNs are not able to explain their reasoning process, therefore limiting their adoption in clinical practice. In this work, we propose an inherently interpretable CNN for regression using similarity-based comparisons (INSightR-Net) and demonstrate our methods on the task of diabetic retinopathy grading. A prototype layer incorporated into the architecture enables visualization of the areas in the image that are most similar to learned prototypes. The final prediction is then intuitively modeled as a mean of prototype labels, weighted by the similarities. We achieved competitive prediction performance with our INSightR-Net compared to a ResNet baseline, showing that it is not necessary to compromise performance for interpretability. Furthermore, we quantified the quality of our explanations using sparsity and diversity, two concepts considered important for a good explanation, and demonstrated the effect of several parameters on the latent space embeddings.