AI-Integrated Clinical Decision Support Framework for Emergency Triage

Abstract

Early patient assessment is essential for effective healthcare delivery; however, the delayed diagnosis, inefficient appointment scheduling, and the absence of real time clinical decision support reduce clinical efficiency and limit timely intervention. This study presents a framework for the Clinical Decision Support System (CDSS) based on machine learning to enable early risk assessment and improve patient prioritization in clinical settings. The proposed system provides a web based graphical user interface through which patients can enter symptoms and relevant health information, allowing it to analyze the data and identify potential early stage respiratory diseases such as influenza, seasonal sore throat infections, and lung cancer. To integrate the best ML model for prediction, three supervised machine learning algorithms:Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Decision Tree were implemented and evaluated using a publicly available lung cancer dataset. Model performance was assessed based on classification accuracy and predictive capability, and the results demonstrated that the Decision Tree model achieved the highest classification performance among the evaluated approaches. The selected model was integrated into the CDSS framework to enable real time risk prediction and automated identification of high-risk patients, assisting healthcare professionals in prioritizing critical cases and supporting timely clinical intervention. In addition, the system architecture allows for future integration of risk-based appointment scheduling to prioritize patients according to predicted disease risk, reduce delays in care, and improve overall clinical workflow efficiency.

Ähnliche Arbeiten