Development of an Ensemble Machine Learning-Based Decision Support System for Patient Admission in Healthcare Unit
Tanusree Pathak, Susmit Sekhar Bhakta, Arghya Kusum Das, Bikash Sadhukhan · 2023
This paper introduces a decision support system designed to enhance patient care in healthcare settings. The system employs various decision-making tools, including alerts for emergencies and recommendations for treatment plans, medications, discharge, or transfer to different units, all based on patient data, and medical history. The system achieves an accuracy of 99% by integrating both knowledge-driven and model-driven approaches, seamlessly incorporating user input such as patient demographics and vital signs, along with real-time data on ICU and general ward bed availability. The output is displayed on a user-friendly website that provides suggestions for immediate action in emergencies based on predefined rules. The experiment involves comparing the performance of several classification algorithms, such as decision tree, logistic regression, support vector machine, naive Bayes and with their ensembles. The results indicate that the ensemble models provide the best outcome. The study highlights the importance of accurate decision-making in healthcare and the need for a reliable decision support system. The study concludes with a future scope for extending the system's capabilities, incorporating additional patient data, and further evaluating the system's effectiveness in real-world healthcare settings. Overall, the proposed system is a well-designed machine learning model that can make quick and accurate decisions based on a patient's vital signs.