Related Experiment Video
Updated: Sep 14, 2025
07:15
Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model
Published on: August 16, 2020
6.9K
Predictive model and risk analysis for outcomes in diabetic foot ulcer using eXtreme Gradient Boosting algorithm and
Lei Gao1, Zi-Xuan Liu2, Jiang-Ning Wang3
1Department of Orthopedics Surgery, Beijing Shijitan Hospital Affiliated to Capital Medical University, Beijing 100038, China.
World Journal of Diabetes
|July 23, 2025
Summary
An interpretable artificial intelligence model using XGBoost effectively predicts diabetic foot ulcer amputation risk. Key factors like Wagner grade 4/5, osteomyelitis, and high C-reactive protein are identified for personalized treatment.
Area of Science:
- Artificial Intelligence in Medicine
- Diabetes Complications Research
- Predictive Modeling in Healthcare
Background:
- Diabetic foot ulcer (DFU) is a severe diabetes complication with high amputation rates and significant societal impact.
- Early identification of risk factors and timely intervention are crucial for reducing DFU-related amputations.
- Advancements in AI offer potential for developing efficient, interpretable predictive models to enhance DFU care.
Purpose of the Study:
- To develop and validate an interpretable predictive model for amputation risk in patients with diabetic foot ulcers.
- To leverage artificial intelligence for improved clinical decision-making in DFU management.
- To identify key risk factors associated with DFU amputation through a data-driven approach.
Main Methods:
- A retrospective study involving 599 DFU patients from Beijing Shijitan Hospital (January 2015 - June 2024).
- Data randomly split into training and testing sets with fivefold cross-validation.
- Three binary variable models built using the eXtreme Gradient Boosting (XGBoost) algorithm, with hyperparameter optimization and SHapley Additive exPlanation (SHAP) for interpretability.
Main Results:
- Out of 599 patients, 157 (26.2%) underwent minor amputation and 50 (8.3%) major amputation.
- All XGBoost models showed good discriminative ability (AUC > 0.7).
- The major amputation prediction model achieved the highest performance (AUC = 0.977), followed by minor (AUC = 0.800) and non-amputation (AUC = 0.772) models. Important predictors included Wagner grade 4/5, osteomyelitis, and elevated C-reactive protein.
Conclusions:
- The XGBoost model demonstrates efficacy in predicting diabetic foot amputation risk.
- The model provides interpretable insights, supporting personalized treatment strategies for DFU patients.
- AI-driven predictive models can significantly aid in managing DFU complications and reducing amputation rates.
Related Concept Videos
01:15Receiver Operating Characteristic Plot
335
A ROC (Receiver Operating Characteristic) plot is a graphical tool used to assess the performance of a binary classification model by illustrating the trade-off between sensitivity (true positive rate) and specificity (false positive rate). By plotting sensitivity against 1 - specificity across various threshold settings, the ROC curve shows how well the model distinguishes between classes, with a curve closer to the top-left corner indicating a more accurate model. The area under the ROC curve...
335
01:23Peripheral Artery Disease V: Postoperative Nursing Management
26
During the postoperative period, it is crucial to focus on maintaining circulation, identifying and managing potential complications, and planning for discharge.Nursing AssessmentVital signs monitoring: Regularly monitor vital signs, including blood pressure, heart rate, respiratory rate, and temperature, to detect early signs of complications such as bleeding and infection.Circulation assessment: Monitor pulses, perform Doppler assessments, and check capillary refill, color, temperature, and...
26