# Opening the black box of deep learning: Validating the statistical association between explainable artificial intelligence (XAI) and clinical domain knowledge in fundus image-based glaucoma diagnosis

arXiv: https://arxiv.org/abs/2504.04549 · HTML: https://mlsanity.com/paper/2504.04549/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 19 of its 67 references that are in the map.
- First posted on arXiv 6 Apr 2025.

## Builds on (most cited first, 19 of 19)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [Learning Deep Features for Discriminative Localization](https://mlsanity.com/paper/1512.04150/) (2015)
- [Score-CAM: Score-Weighted Visual Explanations for Convolutional Neural Networks](https://mlsanity.com/paper/1910.01279/) (2019)
- [Eigen-CAM: Class Activation Map using Principal Components](https://mlsanity.com/paper/2008.00299/) (2020)
- [Axiom-based Grad-CAM: Towards Accurate Visualization and Explanation of CNNs](https://mlsanity.com/paper/2008.02312/) (2020)
- [Weakly Supervised Object Localization and Detection: A Survey](https://mlsanity.com/paper/2104.07918/) (2021)
- [Efficient Medical Image Segmentation Based on Knowledge Distillation](https://mlsanity.com/paper/2108.09987/) (2021)
- [A Comprehensive Analysis of Weakly-Supervised Semantic Segmentation in Different Image Domains](https://mlsanity.com/paper/1912.11186/) (2019)
- [Towards a glaucoma risk index based on simulated hemodynamics from fundus images](https://mlsanity.com/paper/1805.10273/) (2018)
- [Contrastive Counterfactual Visual Explanations With Overdetermination](https://mlsanity.com/paper/2106.14556/) (2021)
- [Simultaneous segmentation and classification of the retinal arteries and veins from color fundus images](https://mlsanity.com/paper/2209.09582/) (2022)
- [Leveraging Anatomical Constraints with Uncertainty for Pneumothorax Segmentation](https://mlsanity.com/paper/2311.15213/) (2023)
- [Clinical Domain Knowledge-Derived Template Improves Post Hoc AI Explanations in Pneumothorax Classification](https://mlsanity.com/paper/2403.18871/) (2024)

## Abstract (arXiv)

While deep learning has exhibited remarkable predictive capabilities in various medical image tasks, its inherent black-box nature has hindered its widespread implementation in real-world healthcare settings. Our objective is to unveil the decision-making processes of deep learning models in the context of glaucoma classification by employing several Class Activation Map (CAM) techniques to generate model focus regions and comparing them with clinical domain knowledge of the anatomical area (optic cup, optic disk, and blood vessels). Four deep neural networks, including VGG-11, ResNet-18, DeiT-Tiny, and Swin Transformer-Tiny, were developed using binary diagnostic labels of glaucoma and five CAM methods (Grad-CAM, XGrad-CAM, Score-CAM, Eigen-CAM, and Layer-CAM) were employed to highlight the model focus area. We applied the paired-sample t-test to compare the percentage of anatomies in the model focus area to the proportion of anatomies in the entire image. After that, Pearson's and Spearman's correlation tests were implemented to examine the relationship between model predictive ability and the percentage of anatomical structures in the model focus area. On five public glaucoma datasets, all deep learning models consistently displayed statistically significantly higher percentages of anatomical structures in the focus area than the proportions of anatomical structures in the entire image. Also, we validated the positive relationship between the percentage of anatomical structures in the focus area and model predictive performance. Our study provides evidence of the convergence of decision logic between deep neural networks and human clinicians through rigorous statistical tests. We anticipate that it can help alleviate clinicians' concerns regarding the trustworthiness of deep learning in healthcare. For reproducibility, the code and dataset have been released at GitHub.