Interpretable and Interactive Deep Multiple Instance Learning for Dental Caries Classification in Bitewing X-rays

TL;DR

This paper introduces an interpretable deep multiple instance learning model for dental caries detection in bitewing X-rays, providing heatmaps and interactive features to improve diagnosis and localization accuracy.

Contribution

It presents a novel, efficient architecture that generates local heatmaps from weak labels and incorporates segmentation guidance for enhanced interpretability and performance.

Findings

Achieves competitive classification accuracy on a large clinical dataset.

Improved localization performance when guided by segmentation models.

Enables human-interpretable heatmaps and interaction for dental radiograph analysis.

Abstract

We propose a simple and efficient image classification architecture based on deep multiple instance learning, and apply it to the challenging task of caries detection in dental radiographs. Technically, our approach contributes in two ways: First, it outputs a heatmap of local patch classification probabilities despite being trained with weak image-level labels. Second, it is amenable to learning from segmentation labels to guide training. In contrast to existing methods, the human user can faithfully interpret predictions and interact with the model to decide which regions to attend to. Experiments are conducted on a large clinical dataset of $\sim$38k bitewings ($\sim$316k teeth), where we achieve competitive performance compared to various baselines. When guided by an external caries segmentation model, a significant improvement in classification and localization performance is observed.