Early Experience With Artificial Intelligence Software to Detect Intracranial Occlusive Stroke in Trauma Patients

Abstract

Objective Identifying ischemic stroke is a diagnostic challenge in the trauma subpopulation. We describe our early experience with artificial intelligence-assisted image analysis software for automatically identifying acute ischemic stroke in trauma patients. Methods Patients were retrospectively screened for (i) admission to the trauma service at a level one trauma center between 2020 and 2022, (ii) radiologist-confirmed intracranial occlusion, (iii) occlusion identified on computed tomography angiography performed within 24 hours of admission, (iv) no intracranial hemorrhage, and (v) contemporaneous analysis with the large vessel occlusion (LVO) detection program. Baseline characteristics, stroke detection, response-activation, and outcome data were summarized. Results Of 9893 trauma patients admitted, 88 (0.89%) patients had a cerebral stroke diagnosis, of which 10 patients (10/88; 11.4%) met inclusion criteria. Most patients were admitted following a fall (8/10; 80%). Six (6/10; 60.0%) patients had LVOs. The program correctly detected 83.3% (5/6) of patients, and these patients were triaged in less than one hour from arrival on average. The program did not falsely identify non-LVOs as LVOs for any patients. Conclusions Identifying adjunct tools to aid timely identification and treatment of ischemic stroke in trauma patients is necessary to increase the chances for meaningful neurological recovery. Our early experience exhibited potential for using automated software to aid occlusion identification and subsequent stroke team mobilization. Future studies in larger cohorts will expand upon these preliminary findings to establish the accuracy and clinical benefit of automated stroke detection tool integration for the trauma population.

Keywords: artificial intelligence; computed tomography angiography; imaging; ischemic stroke; stroke; trauma.

Figure 1. Successful identification by Viz LVO

A 67-year-old female patient presented with left M1 MCA occlusion after a fall. The admission NIHSS score was 15. She was successfully revascularized with thrombectomy. The 24-hour post-intervention NIHSS score was 3. A) Viz LVO identification of M1 occlusion (arrow) on the axial view of CT angiography helical reconstruction. B) Coronal view of non-contrast CT angiography with M1 occlusion (arrow). LVO, large vessel occlusion; MCA, middle cerebral artery; M1, M1 segment of the middle cerebral artery; NIHSS, National Institute of Health Stroke Scale; CT, computed tomography.

Figure 2. Missed identification by Viz LVO

A 47-year-old male patient presented with a left M1 MCA occlusion after motor vehicle collision. The admission NIHSS score was 21. He was successfully revascularized with thrombectomy. The 24-hour post-intervention NIHSS score was 12. A) Viz LVO identification of M1 occlusion (arrow) on axial view of CT angiography helical reconstruction. B) Coronal view of non-contrast CT angiography with M1 occlusion (arrow). LVO, large vessel occlusion; MCA, middle cerebral artery; M1, M1 segment of the middle cerebral artery; NIHSS, National Institutes of Health Stroke Scale; CT, computed tomography.

Figure 3. Occlusion not correctly identified as a large vessel occlusion by Viz LVO

A 86-year-old female patient presented with basilar occlusion after a fall. Admission NIHSS score was 8. She was successfully revascularized with thrombectomy. The 24-hour post-intervention NIHSS score was 1. A) Viz LVO analysis on axial view of CT angiography helical reconstruction. B) Coronal view of non-contrast CT angiography with basilar occlusion (arrow). C) Sagittal view of non-contrast CT angiography with basilar occlusion (arrow). LVO, large vessel occlusion; NIHSS, National Institute of Health Stroke Scale; CT, computed tomography.