Lunit Insight CXR AI Solution Proves Effective in Emergency Patient Classification

Lunit: European Journal of Radiology Publishes Study Highlighting 77% Reduction in Triage Time

Lunit announced on Dec. 26 that a study demonstrating the significant reduction in time required for emergency patient triage using their AI solution, “Lunit Insight CXR,” was published in the European Journal of Radiology. The research, conducted by Dr. Srinivas Sridharan’s team at Changi General Hospital in Singapore, involved the classification of 20,944 chest X-ray images taken in the hospital’s emergency room from August to December last year.

The study categorized the X-ray images into three groups: normal, non-emergency, and emergency, using Lunit’s AI solution. Subsequently, 43 radiologists evaluated the AI’s performance. The results were promising, with the AI demonstrating a sensitivity of 89% and a specificity of 93% for normal cases. For non-emergency cases, the sensitivity was 93% and the specificity was 91%. In emergency cases, the AI achieved a sensitivity of 82% and an impressive specificity of 99%.

One of the key findings of the study was the significant reduction in workload for medical staff when using AI in emergency environments. The average processing time for AI to classify emergency patients was reduced by 77% compared to doctors. In terms of minimum processing time, AI was notably faster, taking just 0.2 seconds compared to 1.7 seconds for doctors.

“This study, designed around the emergency room environment, has verified the effectiveness of AI in quickly and accurately classifying chest X-ray results in various patient groups,” stated the research team. They added, “The results shown by AI will help doctors make immediate decisions about patients.”

Seo Beom-seok, CEO of Lunit, emphasized the clinical usefulness of AI in real-world medical settings. “This study proves the clinical usefulness of AI in actual medical environments,” he said. “In particular, as the Lunit AI solution showed a specificity of 99% in the process of classifying patients requiring emergency care, it will contribute to improving the work efficiency of medical staff in the future.”

The integration of AI in healthcare, particularly in emergency room triage, is a significant advancement. AI technologies, such as machine learning and deep learning, are increasingly being used to enhance diagnostic accuracy, streamline workflows, and improve patient outcomes. In emergency rooms, efficient triage is crucial to ensure that critically ill patients receive immediate attention while optimizing the use of medical resources.

SOURCE: Business Korea