Beyond Labels: Zero-Shot Diabetic Foot Ulcer Wound Segmentation with Self-attention Diffusion Models and the Potential for Text-Guided Customization

2025-04-24

AI Summary

Precise segmentation of diabetic foot ulcers (DFUs) is hindered by severe scarcity of annotated training data. To address this, we propose ADZUS—the first zero-shot, text-guided diffusion-based segmentation model specifically designed for DFUs—requiring no labeled data and generating high-fidelity wound masks in real time solely from clinical text descriptions. Methodologically, ADZUS innovatively integrates a self-attention diffusion architecture with cross-modal text–image alignment and unsupervised feature matching to enable semantic-controllable zero-shot segmentation. Evaluated on a public chronic wound dataset, ADZUS achieves an IoU of 86.68% and precision of 94.69%; on the DFU-specific DSC dataset, it attains a median IoU of 75%, substantially outperforming the supervised baseline FUSegNet (45%). This work breaks the dependency of medical image segmentation on manual annotations and establishes a novel paradigm for intelligent wound assessment in low-resource clinical settings.

Technology Category

• Computer Vision: Segmentation
• Natural Language Processing: Sentence-level Semantics, Textual Inference, etc.
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

Diabetic foot ulcers (DFUs) pose a significant challenge in healthcare, requiring precise and efficient wound assessment to enhance patient outcomes. This study introduces the Attention Diffusion Zero-shot Unsupervised System (ADZUS), a novel text-guided diffusion model that performs wound segmentation without relying on labeled training data. Unlike conventional deep learning models, which require extensive annotation, ADZUS leverages zero-shot learning to dynamically adapt segmentation based on descriptive prompts, offering enhanced flexibility and adaptability in clinical applications. Experimental evaluations demonstrate that ADZUS surpasses traditional and state-of-the-art segmentation models, achieving an IoU of 86.68% and the highest precision of 94.69% on the chronic wound dataset, outperforming supervised approaches such as FUSegNet. Further validation on a custom-curated DFU dataset reinforces its robustness, with ADZUS achieving a median DSC of 75%, significantly surpassing FUSegNet's 45%. The model's text-guided segmentation capability enables real-time customization of segmentation outputs, allowing targeted analysis of wound characteristics based on clinical descriptions. Despite its competitive performance, the computational cost of diffusion-based inference and the need for potential fine-tuning remain areas for future improvement. ADZUS represents a transformative step in wound segmentation, providing a scalable, efficient, and adaptable AI-driven solution for medical imaging.

Problem

• Zero-shot diabetic foot ulcer segmentation without labeled data
• Text-guided customization for dynamic wound analysis
• Overcoming limitations of supervised deep learning models

Innovation

• Zero-shot learning for wound segmentation
• Text-guided diffusion model ADZUS
• Self-attention for dynamic segmentation