UPmémoires

English

Future of traumatic brain triage, using a commercially available AI for the detection of intracranial hemorrhage on brain CT scan : real-life performance study

Backgrounds: Traumatic brain injury is a major cause a morbimortality in the world, and potential intracranial hemorrhage (ICH) is a life-threatening condition that requires rapid diagnosis with computed tomography (CT). Artificial intelligence ICH detection tools are now commercially available.

Purpose: Investigate the real-world performance of qER.AI, an artificial intelligence-based CT hemorrhage detection tool, in a post-traumatic population.

Methods: Retrospective monocentric observationnal study of a dataset of consecutively acquired head CT scans at the emergency radiology unit to explore a brain trauma. AI performance was compared to groundtruth determined by expert consensus. A subset of nighshift cases with radiological report of junior resident was compared to AI results and groundtruth.

Results: 682 head CT scan were analysed. AI demonstrated a sensitivity of 88.8% and specificity of 92.1% overall, with a positive predictive value of 65.4% and a negative predictive value of 98%. AI's performance was comparable to junior residents in detecting ICH, with the latter showing a sensitivity of 85.7% and a notably high specificity of 99.3%. Interestingly, the AI detected two out of three ICH cases missed by junior residents. When AI assistance was integrated, the combined sensitivity improved to 95.2%, and the overall accuracy reached 98.8%, indicating a significant potential for AI to enhance diagnostic accuracy.

Conclusion: This study shows a better performance of AI and radiologist resident associated than each one alone. These results are encouraging to rethink the radiological workflow and the future of triage of this large population of brain traumatised patients in emergency unit.

Keywords : Intracranial hemorrhage, artificial intelligence, traumatic brain injury, emergency radiology, triage.