AI-Based Support System Provides Significant Benefits to Stroke Care

News
Article

Using a cohort of more than 20,000 patients with ischemic stroke, those on AI-CDSS experienced less vascular events of hemorrhagic stroke, myocardial infarction, or vascular death.

Zixiao Li, MD, PhD, an associate professor at Beijing Tiantan Hospital

Zixiao Li, MD, PhD

Findings from the GOLDEN BRIDGE II trial (NCT04524624) showed that the use of an artificial intelligence (AI)-based clinical decision support system, also known as AI-CDSS, had a significantly greater impact on the number of vascular events and stroke care quality than standard care in patients with acute ischemic stroke (AIS). These results were presented as a late-breaker at the 2024 International Stroke Conference (ISC), held February 7-9, in Phoenix, Arizona.1

GOLDEN BRIDGE II was an open-label, cluster-randomized multifaceted intervention study that featured 21,603 patients who received either the AI-CDSS support (n = 11,054) or usual care (n = 10,549). The specific AI-CDSS intervention protocol includes AI-assisted imaging analysis, auxiliary analysis of ischemic stroke etiology and pathogenesis, and guideline-based treatment recommendations for acute and secondary prevention.

A 2-week transition phase of AI-CDSS intensive intervention was completed by neurologists in the intervention hospitals before the patients were enrolled. Results showed that compared with controls, those in the intervention group experienced significantly fewer composite vascular events, defined as ischemic stroke, myocardial infarction, or vascular death (2.9% vs 3.9%; adjusted hazard, 0.75; 95% CI, 0.59-0.95; P = .02) at 3 months.

"The biggest takeaway is that AI can help us increase our performance measures," lead investigator Zixiao Li an associate professor at Beijing Tiantan Hospital, told NeurologyLive®. "We can get the latest trial findings and treatments using AI. This can also save time and allow for us to consider other patient needs."

The study also looked at a composite AIS score, which was defined as the total number of 13 preset performance measures performed divided by the total number of performance measures for which a given patient was eligible. Results revealed that AI-CDSS increased the odds of higher composite score of evidence-based measures than standard care (91.4% vs 89.7%; adjusted OR, 1.20; 95% CI, 1.16-1.25; P <.001). Notably, investigators observed no significant differences in modified Rankin Scale scores of 3-6 between the groups (11.8% vs 9.6%; adjusted OR, 1.24; 95% CI, 0.97-1.59; P = .08) at 3 months.

READ MORE: Adrenomedullin Safe to Treat Ischemic Stroke, May Lead to Improved Outcomes

Following these data, the investigators plan to focus on the performance of AI-CDSS in outpatients. "This includes patients’ adherence to treatments and getting to target secondary prevention level," Li added. "Blood pressure monitoring and blood glucose monitoring can help patients take care of their risk factors as we identify how to use these digital health tools for secondary prevention treatments in the future."

There have been other automated decision-making tools used in the stroke field, including the Computer-Based Decision-Support System for Thrombolysis in Stroke (COMBAT Stroke) and COMPuterized decision Aid for Stroke thrombolysiS (COMPASS) decision aid tool.2,3 COMBAT Stroke, first published in 2013, has been used to assess volumes and ratios of mismatches between perfusion weighted imaging and diffusion weighted imaging. COMPASSS has been used to assist clinicians in making specific clinical decisions about thrombolytic therapy, with a numerical and graphical presentation of the results of risk prediction.

In the published rationale and design for GOLDEN BRIDGE II, Li et al wrote that the application of AI-CDSS in clinical practice faces numerous challenges, including, “how to develop an advanced strategy for the integration of electronic medical records to achieve automatic acquisition and sharing of clinical information; (2) how to construct a standardized clinical knowledge base to provide evidence-based guidelines; and (3) how to combine the output information of AI-CDSS with clinical practice in diagnosis and treatment to establish an effective human–computer interaction."

Click here for more coverage of ISC 2024.

REFERENCES
1. Zhang X, Ding L, Jing J, et al. Effect of an artificial intelligence-based clinical decision support system on stroke care quality and outcomes in patients with acute ischemic stroke (Golden Bridge II): a cluster-randomized clinical trial. Presented at: International Stroke Conference; February 7-9, 2024; Abstract LB15.
2. Nagenthiraja K, Walcott BP, Hansen MB, et al. Automated decision-support system for prediction of treatment responders in acute ischemic stroke. Front Neurol 2013;4:140. doi:10.3389/fneur.2013.00140
3. Flynn D, Nesbitt DJ, Ford GA, et al. Development of a computerised decision aid for thrombolysis in acute stroke care. BMC Med Inform Decis Mak 2015;15:6. doi:10.1186/s12911-014-0127-1
4. Li X, Zhang X, Ding L, et al. Rationale and design of the GOLDEN BRIDGE II: a cluster-randomized multifaceted intervention trial of an artificial intelligence-based cerebrovascular disease clinical decision support system to improve stroke outcomes and care quality in China. Stroke & Vasc Neurol. 2023;svn-2023-002411
Related Videos
Ashkan Shoamanesh, MD
 Jocelyn Y. Cheng, MD
Opeolu Adeoye, MD, MS; Andrew Barerrto, MD, MS
David Z. Rose, MD
Shadi Yaghi, MD, associate professor of neurology at Brown University
© 2024 MJH Life Sciences

All rights reserved.