Clinical Report: DFU-GCNet for Reliable Classification of Diabetic Foot Ulcers

Overview

The DFU-GCNet model achieved a classification accuracy of 97.16% for diabetic foot ulcers, significantly outperforming existing models like VGG16 and EfficientNet. This study emphasizes the importance of explainable AI techniques in enhancing clinical trust and understanding of automated diagnostic systems.

Background

Diabetic foot ulcers (DFUs) are a leading cause of non-traumatic lower limb amputations, significantly impacting patient quality of life. With diabetes prevalence projected to rise, the need for effective and reliable diagnostic tools is critical. Automated systems like DFU-GCNet can improve diagnostic accuracy and reduce healthcare costs, addressing the challenges posed by manual evaluations.

Data Highlights

Key Findings

• DFU-GCNet integrates inception modules with global context blocks for enhanced feature extraction.
• The model provides high-resolution diagnostic heatmaps using explainable AI techniques.
• DFU-GCNet outperformed modern baselines such as VGG16 and EfficientNet in classification tasks.
• Explainable AI methods like GradCAM++ and SHAP enhance the interpretability of the model's decisions.
• The architecture effectively distinguishes between pathological and healthy tissue in DFUs.

Clinical Implications

The DFU-GCNet model can serve as a reliable tool for clinicians in diagnosing diabetic foot ulcers, potentially reducing the risk of amputations. Its explainability features may also foster greater trust in AI-assisted diagnostics among healthcare providers.

Conclusion

DFU-GCNet represents a significant advancement in the automated classification of diabetic foot ulcers, combining high accuracy with interpretability to support clinical decision-making.

Related Resources & Content

1. Frontiers in Endocrinology, 2026 -- Interpretable machine learning for predicting major amputation risk in hospitalized diabetic foot ulcer patients: a single-center study with temporal external validation
2. Frontiers in Medicine, 2026 -- Development of a machine learning-based classification model for diabetic foot in patients with type 2 diabetes: an exploratory analysis with SHAP interpretation
3. Frontiers in Endocrinology, 2026 -- Curative outcomes with metal-containing TCM in diabetic foot ulcers unresponsive to standard therapy: a case series
4. Diabetes Care, 2026 -- Retinopathy, Neuropathy, and Foot Care: Standards of Care in Diabetes—2026
5. IWGDF Guidelines, 2023 -- Offloading guideline (2023 update)
6. Frontiers in Immunology — Expression of lactylation-related genes and their correlation with diabetic foot ulcer occurrence and immune infiltration
7. Telemedical treatment of diabetic foot ulcer in rural and remote areas: a prospective single centre randomised controlled clinical trial
8. 12. Retinopathy, Neuropathy, and Foot Care: Standards of Care in Diabetes—2026 | Diabetes Care | American Diabetes Association
9. Offloading guideline (2023 update) - IWGDF Guidelines