Interpretable Fully Convolutional Classification of Intrapapillary Capillary Loops for Real-Time Detection of Early Squamous Neoplasia

TL;DR

This paper introduces an interpretable deep learning approach using a novel embedded Class Activation Map for real-time detection of early oesophageal squamous neoplasia, improving accuracy and interpretability.

Contribution

It proposes a new deep supervision method and embedded CAM for enhanced interpretability and accuracy in classifying oesophageal tissue in real-time.

Findings

F1-score improved from 87.3% to 92.7%.

Provides detailed attention maps aligned with clinical features.

Enhances interpretability of CNN classification results.

Abstract

In this work, we have concentrated our efforts on the interpretability of classification results coming from a fully convolutional neural network. Motivated by the classification of oesophageal tissue for real-time detection of early squamous neoplasia, the most frequent kind of oesophageal cancer in Asia, we present a new dataset and a novel deep learning method that by means of deep supervision and a newly introduced concept, the embedded Class Activation Map (eCAM), focuses on the interpretability of results as a design constraint of a convolutional network. We present a new approach to visualise attention that aims to give some insights on those areas of the oesophageal tissue that lead a network to conclude that the images belong to a particular class and compare them with those visual features employed by clinicians to produce a clinical diagnosis. In comparison to a baseline method which does not feature deep supervision but provides attention by grafting Class Activation Maps, we improve the F1-score from 87.3% to 92.7% and provide more detailed attention maps.