Prediction of Consultation Wait Time in Outpatient Clinic: An Approach using Neural Network with Optimized Feature Selection

Abstract

Outpatient clinics globally grapple with the uncertainty of patient wait times, a critical factor affecting patient satisfaction. Extended waiting periods are often perceived as a hindrance to timely care, generating stress for both patients and healthcare providers. Accurate prediction of patient wait times can substantially improve patient satisfaction by reducing uncertainty. This study aims to predict patient Consultation Wait Time efficiently in a multispecialty hospital outpatient clinic utilizing a Multilayer Perceptron approach. Feature Selection Methods were employed to enhance the predictive efficiency of the model. The study evaluated various performance metrics, including Accuracy, Recall, Precision, F-measure, and Area Under the Receiver Operating Characteristic Curve (AUC). Temporal features, specifically Visit Time and Consultation Start Time, emerged as significant predictors in the model. Additionally, Vitals Examination was identified as a key factor in predicting Consultation Wait Time. Notably, the model incorporating variables selected through Reciprocal Ranking exhibited robust performance in predicting Consultation Wait Times.