An explainable machine learning model for predicting in‐hospital amputation rate of patients with diabetic foot ulcer

Abstract

Diabetic foot ulcer (DFU) is one of the most serious and alarming diabetic complications, which often leads to high amputation rates in diabetic patients.Machine learning is a part of the field of artificial intelligence, which can automatically learn models from data and better inform clinical decision-making. We aimed to develop an accurate and explainable prediction model to estimate the risk of in-hospital amputation in patients with DFU. A total of 618 hospitalised patients with DFU were included in this study. The patients were divided into non-amputation, minor amputation or major amputation group.Light Gradient Boosting Machine (LightGBM) and 5-fold cross-validation tools were used to construct a multi-class classification model to predict the three outcomes of interest. In addition, we used the SHapley Additive exPlanations (SHAP) algorithm to interpret the predictions of the model. Our area under the receiver-operating-characteristic curve (AUC) demonstrated a 0.90, 0.85 and 0.86 predictive ability for non-amputation, minor amputation and major amputation outcomes, respectively. Taken together, our data demonstrated that the developed explainable machine learning model provided accurate estimates of the amputation rate in patients with DFU during hospitalisation. Besides, the model could inform individualised analyses of the patients' risk factors.Puguang Xie and Yuyao Li contributed equally to this work.

The amputation and mortality of inpatients with diabetic foot ulceration in the COVID‐19 pandemic and postpandemic era: A machine learning study

“…12 In recent years, with the development of artificial intelligence (AI), machine learning algorithms have displayed their advantages in predictions and recognition of the risk for chronic diseases. [13][14][15][16] It is possible to build predictive models of the related diseases by relying on electronic health record (EHR) data. 13 Unfortunately, there are no reports of machine learning models for predicting amputation and mortality in DFU inpatients.…”

“…Although statistical models have been extensively used in the study of chronic diseases such as multiple linear regression and logistic regression analyses, 11 the algorithm is relatively simple, and the results cannot accurately give specific probability values 12 . In recent years, with the development of artificial intelligence (AI), machine learning algorithms have displayed their advantages in predictions and recognition of the risk for chronic diseases 13‐16 . It is possible to build predictive models of the related diseases by relying on electronic health record (EHR) data 13 .…”

Effective prediction model for preventing postoperative deep vein thrombosis during bladder cancer treatment

“…In this study, we selected a 5-fold cross-validation on the basis of the sample size to determine the predictive performance of the model because this is the most common method for solving clinical problems based on machine learning. [10][11][12] In the first step, the data were randomly divided into five parts, one of which was randomly selected as the test set, while the remaining four were used as the training set for model training. In the second step, the process of selecting different test sets and training sets was repeated five times, such that each subset had a chance to be a test set and the rest could be training sets.…”

Impact of acute hyperglycemic crisis episode on survival in individuals with diabetic foot ulcer using a machine learning approach

“…Extreme gradient boosting (XGBoost) algorithm is an efficient and flexible machine learning method with excellent scalability and a high running speed based on gradient tree boosting, which can learn nonlinear, high dimensional relationships from data ( 19 , 20 ). XGBoost often outperforms other machine learning algorithms for prediction with tabular data, and is currently considered as the state-of-the-art method for predicting tabular data ( 21 , 22 ).…”