Managing Patient Flow in Emergency Department
The Challenge
Emergency departments (EDs) provide vital health services to the public, but they often face overcrowding and long waiting times for patients. This can jeopardize patients’ health and impair their quality of care and satisfaction. One of the reasons for this problem is the lack of accurate and timely assessment of patient disease conditions at the triage stage, which is the first point of contact between clinicians and patients. The triage notes contain details of the reason for patients’ visit, including specific symptoms and incidents.
The Solution
We built a decision support system (DSS) that uses AI to analyze the triage notes and detect disease patterns of patients in the early stage of ED. Our DSS uses natural language processing (NLP) and machine learning (ML) techniques to extract relevant information from the triage notes and classify patients into different disease categories. Our DSS also uses a predictive model to estimate the severity and urgency of each patient’s condition and assign them a priority level. Our DSS helps the ED staff to make faster and better decisions about hospital admission and resource allocation.
The Results
The performance of the DSS was tested on thousands of patient records from a medical center emergency department in Detroit, MI to show its capability in performing robust classification that can be used by the hospital surveillance system.