TOM: An Open-Source Tongue Segmentation Method with Multi-Teacher Distillation and Task-Specific Data Augmentation

2025-08-19

AI Summary

Current tongue image segmentation methods suffer from poor robustness and lack of user-friendly tools, hindering the performance of intelligent tongue diagnosis. To address these limitations, we propose TOM—a lightweight tongue segmentation model. First, we introduce a novel multi-teacher knowledge distillation framework that integrates heterogeneous supervision signals. Second, we incorporate a diffusion-model-driven, task-adaptive data augmentation strategy to enhance generalization. Third, we design an extremely compact network architecture achieving 95.22% mIoU while reducing parameter count by 96.6%. Furthermore, we release the first open-source, zero-barrier tongue segmentation tool—freely available and supporting both online and offline deployment. Empirical evaluation demonstrates that TOM significantly improves classification accuracy and interpretability in Traditional Chinese Medicine (TCM) constitutional typing.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Tongue imaging serves as a valuable diagnostic tool, particularly in Traditional Chinese Medicine (TCM). The quality of tongue surface segmentation significantly affects the accuracy of tongue image classification and subsequent diagnosis in intelligent tongue diagnosis systems. However, existing research on tongue image segmentation faces notable limitations, and there is a lack of robust and user-friendly segmentation tools. This paper proposes a tongue image segmentation model (TOM) based on multi-teacher knowledge distillation. By incorporating a novel diffusion-based data augmentation method, we enhanced the generalization ability of the segmentation model while reducing its parameter size. Notably, after reducing the parameter count by 96.6% compared to the teacher models, the student model still achieves an impressive segmentation performance of 95.22% mIoU. Furthermore, we packaged and deployed the trained model as both an online and offline segmentation tool (available at https://itongue.cn/), allowing TCM practitioners and researchers to use it without any programming experience. We also present a case study on TCM constitution classification using segmented tongue patches. Experimental results demonstrate that training with tongue patches yields higher classification performance and better interpretability than original tongue images. To our knowledge, this is the first open-source and freely available tongue image segmentation tool.

Problem

• Develops open-source tongue segmentation tool for medical diagnosis
• Addresses lack of robust tongue image segmentation methods
• Reduces model size while maintaining high segmentation accuracy

Innovation

• Multi-teacher knowledge distillation model
• Diffusion-based data augmentation method
• Deployed online-offline segmentation tool