Deep Modeling and Interpretation for Bladder Cancer Classification

TL;DR

This study evaluates deep learning models, including CNNs and transformers, for bladder cancer classification, focusing on accuracy, calibration, and interpretability, revealing that different models excel in different scenarios.

Contribution

It provides a comprehensive comparison of 13 deep models for bladder cancer classification, analyzing their calibration and interpretability, which is novel in medical imaging context.

Findings

ConvNext models have limited generalization (~60% accuracy)

ViTs show better calibration than CNNs

Different models are suitable for in-distribution and out-of-distribution samples

Abstract

Deep models based on vision transformer (ViT) and convolutional neural network (CNN) have demonstrated remarkable performance on natural datasets. However, these models may not be similar in medical imaging, where abnormal regions cover only a small portion of the image. This challenge motivates this study to investigate the latest deep models for bladder cancer classification tasks. We propose the following to evaluate these deep models: 1) standard classification using 13 models (four CNNs and eight transormer-based models), 2) calibration analysis to examine if these models are well calibrated for bladder cancer classification, and 3) we use GradCAM++ to evaluate the interpretability of these models for clinical diagnosis. We simulate $\sim 300$ experiments on a publicly multicenter bladder cancer dataset, and the experimental results demonstrate that the ConvNext series indicate limited generalization ability to classify bladder cancer images (e.g., $\sim 60\%$ accuracy). In addition, ViTs show better calibration effects compared to ConvNext and swin transformer series. We also involve test time augmentation to improve the models interpretability. Finally, no model provides a one-size-fits-all solution for a feasible interpretable model. ConvNext series are suitable for in-distribution samples, while ViT and its variants are suitable for interpreting out-of-distribution samples.