Weakly-supervised segmentation using inherently-explainable classification models and their application to brain tumour classification
This work addresses the challenge of building clinical trust in AI for brain tumour classification by providing interpretability without extensive annotations, though it is incremental as it builds on existing methods with a novel integration.
The paper tackled the problem of opacity in deep learning models for medical imaging and the need for pixel-wise annotations in segmentation by introducing inherently explainable classifiers that generate localisation heatmaps for weakly-supervised segmentation, achieving a peak F1-score of 0.93 for classification and a median Dice score of 0.728 for segmentation.
Deep learning has demonstrated significant potential in medical imaging; however, the opacity of "black-box" models hinders clinical trust, while segmentation tasks typically necessitate labourious, hard-to-obtain pixel-wise annotations. To address these challenges simultaneously, this paper introduces a framework for three inherently explainable classifiers (GP-UNet, GP-ShuffleUNet, and GP-ReconResNet). By integrating a global pooling mechanism, these networks generate localisation heatmaps that directly influence classification decisions, offering inherent interpretability without relying on potentially unreliable post-hoc methods. These heatmaps are subsequently thresholded to achieve weakly-supervised segmentation, requiring only image-level classification labels for training. Validated on two datasets for multi-class brain tumour classification, the proposed models achieved a peak F1-score of 0.93. For the weakly-supervised segmentation task, a median Dice score of 0.728 (95% CI 0.715-0.739) was recorded. Notably, on a subset of tumour-only images, the best model achieved an accuracy of 98.7%, outperforming state-of-the-art glioma grading binary classifiers. Furthermore, comparative Precision-Recall analysis validated the framework's robustness against severe class imbalance, establishing a direct correlation between diagnostic confidence and segmentation fidelity. These results demonstrate that the proposed framework successfully combines high diagnostic accuracy with essential transparency, offering a promising direction for trustworthy clinical decision support. Code is available on GitHub: https://github.com/soumickmj/GPModels