News Details

In a groundbreaking study spearheaded by Mr. Helal Uddin, Senior Lecturer, Department of Sociology, East West University, and his research team, the performance of ChatGPT in mass casualty incident triage was assessed before and after instructional input. Collaborating with Rick KyeGan, Ann ZeeGan, YingYingYew, and Pedro Arcos González, the research aimed to gauge ChatGPT's transformative impact on AI in emergency scenarios.

Following the teaching of the Simple Triage And Rapid Treatment (START) algorithm, ChatGPT demonstrated an impressive 80% overall triage accuracy, with only 20% over-triage cases. This surpassed the mean accuracy of medical students, which stood at 64.3%.

Helal Uddin's leadership played a pivotal role in the success of the study. Qualitative analysis focused on key themes related to mass casualty incidents, revealing consistent performance in 'walking-wounded,' 'respiration,' 'perfusion,' and 'mental status' both before and after START triage teaching.

The thematic analysis uncovered additional insights into ChatGPT's capabilities, including 'disclaimer,' 'prediction,' 'management plan,' and 'assumption.' These findings highlight ChatGPT's potential in effectively responding to mass casualty incident questionnaires, with Helal Uddin's leadership contributing significantly to the study's success. The research marks a notable stride in advancing AI applications for critical emergency scenarios.