Paediatric upper limb fracture healing time prediction using a machine learning approach

To analyse and predict the healing time of upper limb fractures in children, machine learning algorithms were used. Paediatric orthopaedic data was obtained from University Malaya Medical Centre. The set includes radiographs involving radius, ulna, humerus children under age twelve, with ages recorded date initial injury. Inputs assessment included: age, gender, bone type, number bones involved, fracture angulation distance fracture. Random Forest (RF) Support Vector Regression (SVR) used to identify variables associated time. Self Organizing Maps then for visualization ordination factors Algorithms performance measured using root mean square error (RMSE). A significant determinant part, angulation, distance. Wilcoxon signed ranked test reported there is a difference between prediction result SVR model (RMSE = 2.56) RF 2.66). Predicting can be treatment process follow up period general practitioners medical officers. algorithm deployed online at https://kidsfractureexpert.com/.