MediQueue: An ML-Driven Hospital Queue Management System with Real-Time Wait Time Prediction

Purpose: Hospital out-patient departments (OPDs) in India face severe queue inefficiencies with average waiting times of 90+ minutes and poor patient communication. Methodology: This study presents MediQueue — a full-stack intelligent queue management system built with React.js, Node.js, MySQL, Socket.io, and a self-learning ML engine. A dual-prediction architecture (Random Forest + equal-distribution fallback) predicts per-department wait times. A nightly recalibration scheduler updates slot capacities from verified treatment records. Findings: The system achieves a Mean Absolute Error (MAE) of 2.3 minutes after accumulating five verified samples per department. All role dashboards (patient, doctor, admin) show identical wait time estimates using the equal-distribution formula. Conclusion: MediQueue demonstrates that a self-bootstrapping ML system — requiring no pre-labelled dataset — can significantly improve OPD efficiency, patient communication, and clinical workflow management.