Machine learning constructs a diagnostic prediction model for gangrenous perforation of acute appendicitis in elderly patients

Background As life expectancy rises and elderly populations grow, acute appendicitis incidence increases, often manifesting with nonspecific symptoms that challenge diagnosis. This study applied machine learning techniques to build a predictive model for gangrenous perforation, examining clinical features and risk factors in elderly patients with acute appendicitis. Methods We conducted a retrospective analysis of elderly patients undergoing laparoscopic appendectomy for acute appendicitis at The Second Affiliated Hospital of Kunming Medical University, China, from June 2021 to January 2024 (n = 251). Patients were classified into gangrenous perforation (n = 69) and non-gangrenous (n = 182) groups, then randomly split into training (70%) and test (30%) sets. Univariate analyses, including t-tests, Spearman correlations, and chi-square tests, assessed differences across 38 variables in both sets. The least absolute shrinkage and selection operator (LASSO) screened features from the training set, informing models via logistic regression (LR), extreme gradient boosting (XGBoost), support vector machines (SVM), and random forest (RF). Model performance was evaluated using area under the receiver operating characteristic curve (AUC), with decision curve analysis assessing clinical applicability. Results In the training set, RF yielded the highest AUC (0.999, 95% CI: 0.998–1.000), followed by XGBoost (0.975, 95% CI: 0.953–0.998), LR (0.774, 95% CI: 0.692–0.856), and SVM (0.768, 95% CI: 0.684–0.852). In the test set, LR performed best (AUC 0.768, 95% CI: 0.642–0.893), surpassing SVM (0.751, 95% CI: 0.620–0.882), XGBoost (0.725, 95% CI: 0.584–0.867), and RF (0.686, 95% CI: 0.533–0.839). LR also showed the highest accuracy (0.784) and specificity (0.741), while XGBoost had the greatest sensitivity (0.836). Conclusions Among the models, LR emerged as the most effective for predicting gangrenous perforation in elderly acute appendicitis patients, offering robust accuracy and reliability. Its nomogram provides a noninvasive aid for clinical diagnosis.