# A Graph-Augmented knowledge Distillation based Dual-Stream Vision Transformer with Region-Aware Attention for Gastrointestinal Disease Classification with Explainable AI

arXiv: https://arxiv.org/abs/2512.21372 · HTML: https://mlsanity.com/paper/2512.21372/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Distillation for medical imaging, one of 12 directions in Knowledge distillation.
- Builds on 4 of its 10 references that are in the map.
- First posted on arXiv 24 Dec 2025.

## Placed in

- Knowledge distillation / [Distillation for medical imaging](https://mlsanity.com/distill/distillation-for-medical-imaging/)

## Builds on (most cited first, 4 of 4)

- [DeepSMILE: Contrastive self-supervised pre-training benefits MSI and HRD classification directly from H&E whole-slide images in colorectal and breast cancer](https://mlsanity.com/paper/2107.09405/) (2021)
- [EndoOOD: Uncertainty-aware Out-of-distribution Detection in Capsule Endoscopy Diagnosis](https://mlsanity.com/paper/2402.11476/) (2024)
- [An Anatomy-aware Framework for Automatic Segmentation of Parotid Tumor from Multimodal MRI](https://mlsanity.com/paper/2210.01467/) (2022)
- [Deep Transfer Learning for Texture Classification in Colorectal Cancer Histology](https://mlsanity.com/paper/2004.01614/) (2020)

## Abstract (arXiv)

The accurate classification of gastrointestinal diseases from endoscopic and histopathological imagery remains a significant challenge in medical diagnostics, mainly due to the vast data volume and subtle variation in inter-class visuals. This study presents a hybrid dual-stream deep learning framework built on teacher-student knowledge distillation, where a high-capacity teacher model integrates the global contextual reasoning of a Swin Transformer with the local fine-grained feature extraction of a Vision Transformer. The student network was implemented as a compact Tiny-ViT structure that inherits the teacher's semantic and morphological knowledge via soft-label distillation, achieving a balance between efficiency and diagnostic accuracy. Two carefully curated Wireless Capsule Endoscopy datasets, encompassing major GI disease classes, were employed to ensure balanced representation and prevent inter-sample bias. The proposed framework achieved remarkable performance with accuracies of 0.9978 and 0.9928 on Dataset 1 and Dataset 2 respectively, and an average AUC of 1.0000, signifying near-perfect discriminative capability. Interpretability analyses using Grad-CAM, LIME, and Score-CAM confirmed that the model's predictions were grounded in clinically significant tissue regions and pathologically relevant morphological cues, validating the framework's transparency and reliability. The Tiny-ViT demonstrated diagnostic performance with reduced computational complexity comparable to its transformer-based teacher while delivering faster inference, making it suitable for resource-constrained clinical environments. Overall, the proposed framework provides a robust, interpretable, and scalable solution for AI-assisted GI disease diagnosis, paving the way toward future intelligent endoscopic screening that is compatible with clinical practicality.