TTE-CAM: Built-in Class Activation Maps for Test-Time Explainability in Pretrained Black-Box CNNs

TL;DR

TTE-CAM is a novel test-time framework that transforms pretrained black-box CNNs into self-explainable models, maintaining high accuracy while providing faithful built-in explanations for medical image analysis.

Contribution

It introduces a convolution-based replacement of the classification head that preserves performance and offers faithful explanations, bridging the gap between post-hoc and inherently interpretable models.

Findings

Preserves original CNN predictive performance.

Provides built-in explanations comparable to post-hoc methods.

Code available at https://github.com/kdjoumessi/Test-Time-Explainability

Abstract

Convolutional neural networks (CNNs) achieve state-of-the-art performance in medical image analysis yet remain opaque, limiting adoption in high-stakes clinical settings. Existing approaches face a fundamental trade-off: post-hoc methods provide unfaithful approximate explanations, while inherently interpretable architectures are faithful but often sacrifice predictive performance. We introduce TTE-CAM, a test-time framework that bridges this gap by converting pretrained black-box CNNs into self-explainable models via a convolution-based replacement of their classification head, initialized from the original weights. The resulting model preserves black-box predictive performance while delivering built-in faithful explanations competitive with post-hoc methods, both qualitatively and quantitatively. The code is available at https://github.com/kdjoumessi/Test-Time-Explainability