Explainable Multilayer Neural network for prediction of Lower limb amputation in hospitalized diabetic foot patients

Background and Aims: Diabetic foot ulcers (DFUs) account for over ۸۰% of non-traumatic lower limb amputations globally. Early risk prediction is critical for preventing amputations, but traditional systems like PEDIS lack precision and interpretability. Explainable AI models offer a promising solution by combining accuracy with clinical transparency. This study aims to develop an interpretable neural network for predicting amputation risk in hospitalized DFU patients, identifying actionable biomarkers. Methods: A retrospective cohort of ۲۰۳ hospitalized diabetic foot patients (۷۰ instances of lower limb amputation) from Sina hospital, Tabriz, Iran in ۲۰۲۳ was analyzed. Ethical approval was obtained from IRB (IR.TBZMED.REC.۱۳۹۹.۵۷۱). Twenty-one features were selected via ReliefF analysis. Numerical features were standardized using z-scores. A multilayer perceptron (۲۱-۴۰-۱۸-۲) with Tanh activation in hidden layers, Adam optimizer (learning rate=۰.۰۰۱), and L۲ regularization (α=۰.۰۰۱) was modeled. The samples were randomly divided into train and test sets with a ratio of ۸:۲. SHAP values were computed to interpret feature contributions. Models were implemented in python ۳.۱۱.۸, Orange ۳.۳۷, using appropriate libraries. The PEDIS scoring system was used as comparison with a cutoff of ≥۷. Results: The neural network achieved an AUC of ۰.۹۶۲, sensitivity=۷۸.۶%, specificity=۹۲.۳%, and F۱=۰.۸۱۵, and Accuracy=۸۷.۵%. In comparison, PEDIS score achieved an AUC of ۰.۸۸۱, sensitivity of ۷۸.۲۰%, and specificity of ۸۲.۸۶%. SHAP analysis identified Wagner ۴ ulcer, ulcer size ≥۳cm, Ulcer depth of bone or joint, greater ESR values, absence of prior amputation, and higher platelet counts as top predictors. Conclusion: Our explainable AI model demonstrated clinically significant accuracy in predicting amputation risk, outperforming PEDIS while providing interpretable insights into ulcer severity and inflammatory markers. High specificity reduces unnecessary interventions, addressing a key concern in DFU management. Future work should validate the model across larger, more diverse cohorts and integrate emerging biomarkers (e.g., proteomics) for broader clinical adoption. This study highlights the transformative potential of explainable AI in high-stakes diabetic care.

کلیدواژه ها:

Artificial intelligence ، Medicine ، Diabetic foot ، Amputation