Publication

Abstract : Children’s appendicitis is often diagnosed using outdated techniques that may not be as quick or accurate as they need to be. The present investigation employs sophisticated machine learning (ML) methodologies to mitigate these constraints and enhance the diagnostic results. We investigated the effectiveness of many ensemble machine learning (ML) models, such as Gradient Boosting Machine (GBM), CatBoost, LightGBM, AdaBoost, and Decision Trees, using a proprietary dataset that contains a diverse range of pediatric appendicitis cases. After extensive training and validation, these models showed amazing accuracy and precision rates—the best models reaching up to 99% accuracy. In addition to outlining each model’s effectiveness, our comparison study clarifies which models are applicable in clinical contexts. This research work addresses how these models might improve clinical processes, lower diagnostic mistake rates, and speed up and improve the accuracy of clinical decisionmaking. The findings of this study demonstrate the revolutionary potential of machine learning to improve pediatric healthcare by offering reliable instruments for intricate medical diagnosis.