In accordance with a preliminary study of late stroke, ischemic stroke survivors who received care recommendations from a synthetic intelligence (AI)-based system had fewer recurrent strokes, heart attacks or vascular deaths over three months compared with individuals who didn’t receive stroke treatment. was based on artificial intelligence tools. science presented today on the American Stroke Association’s 2024 International Stroke Conference. The meeting, held in person in Phoenix on February 7–9, 2024, is the world’s most significant meeting for stroke and brain health researchers and clinicians.

This study showed that a synthetic intelligence-based clinical decision support system for stroke treatment was effective and feasible in a clinical setting in China and improved patient outcomes. This kind of technology helps neuroscientists by facilitating the exchange of data between humans and artificial intelligence, leveraging their combined strengths.”

Zixiao Li, MD, lead creator of the study, chief medical officer, professor and deputy director of neurology at Beijing Tiantan Hospital in Beijing, China

In accordance with Li, ischemic stroke is the leading reason for death in China. Timely assessment and decisions regarding stroke diagnosis and treatment are crucial to restoring blood flow and minimizing the quantity of brain damage. In accordance with the most recent data published in 2024 Heart Disease and Stroke Statistics: A Report of US and Global Data From the American Heart Association, there have been 7.44 million stroke-related deaths worldwide in 2021, of which about half were ischemic stroke . Within the U.S., 87% of strokes are ischemic, which occur when blood vessels within the brain narrow or turn into clogged with plaque, cutting off blood flow to the brain.

In a clinical trial called GOLDEN BRIDGE II, 77 hospitals in China were randomly assigned to perform diagnosis and treatment of ischemic stroke patients based on the recommendations of the substitute intelligence technology system or the assessments and proposals issued by the hospital’s stroke management team. The AI ​​system integrated participants’ AI-interpreted brain imaging scans with established clinical knowledge of stroke diagnosis, stroke classification, and guideline-recommended treatment and secondary stroke prevention strategies.

The researchers then measured the variety of vascular events in greater than 20,000 study participants; ischemic strokes, hemorrhagic strokes, heart attacks or death from a vascular event -; amongst all study participants after the primary ischemic stroke in the course of the three-month remark period.

The evaluation showed:

• Using an AI-based clinical decision support system reduced the chance of recent vascular events by 25.6% within the three months after the primary stroke and likewise improved the standard of stroke care, leading to patients being more more likely to be treated in accordance with guidelines.
• After three months, participants treated in hospitals using AI experienced fewer total vascular events in comparison with those receiving standard stroke assessment and treatment (2.9% vs. 3.9%).
• There have been no statistically significant differences in the extent of physical disability between patients receiving AI-guided care or the usual care group after three months, as assessed by the modified Rankin scale -; a tool for determining the degree of disability in individuals who have had a stroke.

“The reduction within the number of recent vascular events is a big finding since it shows that artificial intelligence can truly change the best way we treat stroke and profit a big population of stroke survivors,” said Li, who can be a professor on the China National Clinical Center for Research on Neurological Diseases ; Artificial Intelligence in Cerebrovascular Diseases Research Unit, Chinese Academy of Medical Sciences; and the China Brain Research Institute, all in Beijing.

“We hope that in the long run, more applications of artificial intelligence shall be validated through clinical trials and that the clinical decision support system might be expanded to cover more points of stroke care, including reperfusion therapy and long-term secondary prevention, rehabilitation, and so forth. At the identical time, we also hope that the applications of AI might be expanded to use to other conditions.”

Details and background of the study:

• The study was conducted from January 2021 to June 2023 in 77 hospitals in various regions of China.
• A complete of 21,603 adults hospitalized with acute ischemic stroke were included within the study; roughly one third were women and the common age of participants was 67 years.
• The variety of participants included in each treatment approach was almost the identical: 11,054 participants received AI-based assessment and treatment, and 10,549 received usual care and treatment based on assessments and proposals from hospital neurologists.
• Just about all participants (21,579) were included in the ultimate data assessment because they accomplished the three-month follow-up period.
• Neurologists at hospitals testing the AI ​​technology accomplished training in an AI clinical decision support system for stroke before the beginning of the study.
• The standard of stroke care was measured using the internationally recognized composite scale of evidence-based quality of care indicators for acute ischemic stroke patients, including eight indicators initially of hospitalization and five indicators at discharge, Li said.

Limitations of the study include that hospitals were randomized to either an AI-based strategy or standard care, relatively than individual patients; and differences in care patterns and outcomes between inpatient and subsequent outpatient care could have influenced the outcomes. Moreover, whether improvements in care and outcomes might be sustained require further evaluation, and the functionality of an AI-based clinical decision support system for stroke may should be continually updated using revised evidence-based clinical guidelines. More comprehensive and sustainable models for the clinical application of an AI-based clinical decision support system for stroke must be explored for other conditions and in other countries.