Enhanced Neonatal Surgical Site Infection Prediction Model Utilizing Statistically and Clinically Significant Variables in Combination with a Machine Learning Algorithm

ConclusionsThe hybrid model had similar predictability as other models with fewer and more clinically relevant variables. Machine-learning algorithms can identify important novel characteristics, which enhance clinical prediction models.SummaryThis study evaluated risk factors associated with neonatal surgical site infection (SSI) utilizing multiple logistic regression and a random forest classifier. Operative time, open surgical technique, and preoperative supplemental nutrition were associated with SSI. A hybrid multiple logistic regression model was developed based on the random forest and clinical knowledge, and predicted neonatal SSI as well as the other models while being more feasible.
Source: The American Journal of Surgery - Category: Surgery Source Type: research