Promising AI-based approach to early detection of suicide-related behavior at emergency rooms
Researchers from the University of Johannesburg (UJ) have published a promising approach to identifying people at risk of suicide-related behavior (SRB) in the journal Scientific Reports.
South African researchers at the University of Johannesburg have developed an AI-based system to identify patients at risk of suicide-related behavior (SRB) in emergency room (ER) settings. The AI model, built using historical U.S. patient data, analyzes routine intake information and vital signs to estimate SRB risk. Assistant lecturer Tsholofelo Mokheleli, the lead researcher, explains that the model is intended to assist ER staff, especially during busy shifts when one staff member may have to triage 30 or more patients.
The model flags patients at risk and provides reasons, but clinicians still need to make the final decision on the severity of the risk. The study's findings indicate that the AI model correctly predicted 866 patients as not at risk per 1,000 ER visits, while incorrectly identifying 13 as not at risk despite a documented SRB outcome within 30 days.
The model also flagged 121 patients as potentially at risk, a task that would be resource-intensive for clinicians to handle individually.
Written by urgent.news from Medical Xpress's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
