Patchnet: Interpretable Neural Networks for Image Classification

TL;DR

PatchNet is an interpretable neural network model that highlights class-specific features in images, balancing global context and accuracy, and is validated on skin lesion classification with clear visual explanations.

Contribution

It introduces PatchNet, a novel method for generating interpretable visual heatmaps for image classification, with a mathematical analysis of its tradeoffs and empirical validation.

Findings

Produces sharp visual heatmaps of learned features

Quantitatively aligns with domain expert features

Effective in skin lesion classification

Abstract

Understanding how a complex machine learning model makes a classification decision is essential for its acceptance in sensitive areas such as health care. Towards this end, we present PatchNet, a method that provides the features indicative of each class in an image using a tradeoff between restricting global image context and classification error. We mathematically analyze this tradeoff, demonstrate Patchnet's ability to construct sharp visual heatmap representations of the learned features, and quantitatively compare these features with features selected by domain experts by applying PatchNet to the classification of benign/malignant skin lesions from the ISBI-ISIC 2017 melanoma classification challenge.