Automated Large Vessel Occlusion Detection Software and Thrombectomy Treatment Times: A Cluster Randomized Clinical Trial

The Role of Artificial Intelligence-Powered Imaging in Cerebrovascular Accident Detection

Cerebrovascular accidents (CVAs) often occur suddenly and abruptly, leaving patients with long-lasting disabilities that place a huge emotional and economic burden on everyone involved. CVAs result when emboli or thrombi travel to the brain and impede blood flow; the subsequent lack of oxygen supply leads to ischemia and eventually tissue infarction. The most important factor determining the prognosis of CVA patients is time, specifically the time from the onset of disease to treatment. Artificial intelligence (AI)-assisted neuroimaging alleviates the time constraints of analysis faced using traditional diagnostic imaging modalities, thus shortening the time from diagnosis to treatment. Numerous recent studies support the increased accuracy and processing capabilities of AI-assisted imaging modalities. However, the learning curve is steep, and huge barriers still exist preventing a full-scale implementation of this technology. Thus, the potential for AI to revolutionize medicine and healthcare delivery demands attention. This paper aims to elucidate the progress of AI-powered imaging in CVA diagnosis while considering traditional imaging techniques and suggesting methods to overcome adoption barriers in the hope that AI-assisted neuroimaging will be considered normal practice in the near future. There are multiple modalities for AI neuroimaging, all of which require collecting sufficient data to establish inclusive, accurate, and uniform detection platforms. Future efforts must focus on developing methods for data harmonization and standardization. Furthermore, transparency in the explainability of these technologies needs to be established to facilitate trust between physicians and AI-powered technology. This necessitates considerable resources, both financial and expertise wise which are not available everywhere.

The clinician-AI interface: intended use and explainability in FDA-cleared AI devices for medical image interpretation

As applications of AI in medicine continue to expand, there is an increasing focus on integration into clinical practice. An underappreciated aspect of this clinical translation is where the AI fits into the clinical workflow, and in turn, the outputs generated by the AI to facilitate clinician interaction in this workflow. For instance, in the canonical use case of AI for medical image interpretation, the AI could prioritize cases before clinician review or even autonomously interpret the images without clinician review. A related aspect is explainability - does the AI generate outputs to help explain its predictions to clinicians? While many clinical AI workflows and explainability techniques have been proposed, a summative assessment of the current scope in clinical practice is lacking. Here, we evaluate the current state of FDA-cleared AI devices for medical image interpretation assistance in terms of intended clinical use, outputs generated, and types of explainability offered. We create a curated database focused on these aspects of the clinician-AI interface, where we find a high frequency of "triage" devices, notable variability in output characteristics across products, and often limited explainability of AI predictions. Altogether, we aim to increase transparency of the current landscape of the clinician-AI interface and highlight the need to rigorously assess which strategies ultimately lead to the best clinical outcomes.

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

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.

The Influence of the Novel Computer-Aided Triage System Based on Artificial Intelligence on Endovascular Therapy in Patients with Large Vascular Occlusions: A Meta-Analysis

OBJECTIVE

The integration of artificial intelligence (AI) with modern healthcare has become increasingly prominent. The purpose of this study is to explore the impact of the novel computer-aided triage system based on artificial intelligence (AI-CTS) on endovascular therapy (EVT) in patients with large vascular occlusions (LVO). This study marks the first comprehensive systematic review and meta-analysis on the subject.

METHODS

A comprehensive study was performed on PubMed, Medline, Embase, Cochrane Library, and Chinese databases from their establishment to September 2023, in accordance with PRISMA recommendations. RevMan 5.4 software was used for summative analysis. The outcomes included door-to-groin (DTG) time, time from CT scan initiation to EVT, time from CT scan to reperfusion, and 90-day modified Rankin Scale (mRS).

RESULTS

A total of 7 studies involving 752 participants were included in the meta-analysis. The pooled results demonstrated that patients in the post-AI group had less time of DTG [SMD, 0.54; 95% CI, 0.40-0.69; P < 0.00001] and CT scan to EVT [SMD, 0.57; 95% CI, 0.42-0.73; P < 0.00001], as well as less time of CTA to recanalization [SMD, 0.63; 95% CI, 0.36-0.90; P < 0.00001]. There was no significant difference between the 2 groups in terms of the mRS at 90 days [OR, 0.66; 95% CI, 0.43-1.01; P = 0.06].

CONCLUSIONS

The combination of AI-CTS and EVT has improved the therapy process for LVO patients. However, the improvement in mRS at 90 days was not significant; further research is warranted.