Darren Wang | December 3rd, 2024
Stroke is a leading cause of long-term disability in adults across the globe, resulting in substantial economic and social burdens on families and societies. Stroke occurs when the blood supply to the brain is disrupted, either due to a blockage (ischemic stroke) or a ruptured blood vessel (hemorrhagic stroke). In the United States alone, 795,000 people suffer from strokes annually, of which 137,000 people die.
Current stroke detection methods
Current clinical stroke detection methods involve a series of tests and imaging techniques to diagnose and assess the extent of a potential stroke.
The physical exam includes heart function and blood pressure monitoring along with a neurological exam to evaluate nervous system function. Blood tests are conducted to assess clotting speed, blood sugar levels, and the presence of infections. In addition, several types of imaging are used to diagnose strokes. Computerized tomography (CT) scans use X-rays to create detailed brain images. Magnetic resonance imaging (MRI) uses radio waves and a magnetic field to provide a comprehensive view of brain tissues and detect stroke-related damage. Carotid ultrasound examines the carotid arteries in the neck to identify fatty deposits and blood flow. Cerebral angiograms, although less common, involve inserting a catheter to visualize blood vessels in the brain and neck using X-ray imaging. Echocardiograms use sound waves to capture detailed images of the heart.
Novel advancements in stroke detection
In recent years, because of the rapid technological advancements in healthcare, there have been various novel uses of AI in stroke-related cases. For example, group of researchers from Youngsan University in Korea proposed an AI-based stroke prediction system using real time muscle electrical signals. There have also been numerous recent attempts to implement computer vision — a field of AI that enables machines to interpret and analyze visual information — into brain imaging.
The use of AI computer vision was also gradually introduced into the field of stroke detection. For instance, StrokeSave offers a mobile platform that utilizes real-time vascular patient data along with deep learning (a type of AI that uses multi-layered neural networks to automatically learn patterns from vast amounts of data) and computer vision. Together, these technologies produce efficient and accurate diagnoses by conducting machine learning analyses of common stroke symptoms, including high blood pressure, irregular heart rhythms, hypoxia, facial paralysis, vocal paralysis, and hypertensive retinopathy.
Stroke intervention at Vanderbilt
Researchers at Vanderbilt University are developing a new catheter technology to improve the efficiency and safety of stroke treatments. In their method, a thin flexible catheter is inserted into an artery located in the patient’s groin area, and the surgeon manipulates it towards the suspected location of the clot by observing X-ray visuals displayed on a monitor. The technology allows for better navigation to the blood clot, which attempts to reduce the time and complications associated with the current methods.
Just two years ago, Vanderbilt Assistant Professor Dr. James Weiner’s Neuralert was recognized as one of Time’s Best Inventions of 2022. Neuralert is a wristband device that detects strokes early by monitoring asymmetric arm movement, which is a common stroke indicator. Recognized by the FDA as a Breakthrough Device, Neuralert can provide rapid alerts, even when the patient is asleep. Dr. Weimer’s research is centered on creating technology for medical use that provides actionable feedback to healthcare providers.
The landscape of stroke diagnosis and treatment is evolving rapidly and has promising new technologies. Advancements such as AI-based stroke prediction, computer vision in brain imaging, and innovative medical devices such as Neuralert are transforming the healthcare industry’s approach to stroke care. The progress made at institutions like Vanderbilt is a testament to the relentless pursuit of medical innovation worldwide.
As we inch closer and closer towards autonomous stroke detection technologies that are more accurate, more time-effective, and safer, we pave the way for a transformative shift in medical care. This shift is one that scientists hope will drastically reduce the long-term disability associated with strokes and significantly improve the quality of life for millions of patients globally.
References
Gupta, A. (2019, July 9). Strokesave: A novel, high-performance mobile application for stroke diagnosis using Deep Learning and Computer Vision. arXiv.org. https://arxiv.org/abs/1907.05358
Hemorrhagic stroke. www.stroke.org. https://www.stroke.org/en/about-stroke/types-of-stroke/hemorrhagic-strokes-bleeds
Ischemic stroke (clots). www.stroke.org. https://www.stroke.org/en/about-stroke/types-of-stroke/ischemic-stroke-clots#:~:text=An%20ischemic%20stroke%20occurs%20when,that%20line%20the%20vessel%20walls.
Johnson, L. (1970, May 9). Vanderbilt researchers’ novel catheter-based technology to make endovascular procedures more efficient and safe. Vanderbilt University. https://engineering.vanderbilt.edu/2024/05/09/vanderbilt-researchers-novel-catheter-based-technology-to-make-endovascular-procedures-more-efficient-and-safe/
Life-saving stroke detection. Neuralert. (2024, August 29). https://neuralert.co/
Shapiro, M. (1970, December 5). Two Vanderbilt faculty win “time” best inventions of 2022. Vanderbilt University. https://news.vanderbilt.edu/2022/12/05/two-vanderbilt-faculty-win-time-best-inventions-of-2022/
Soun, J. E., Zolyan, A., McLouth, J., Elstrott, S., Nagamine, M., Liang, C., Dehkordi-Vakil, F. H., Chu, E., Floriolli, D., Kuoy, E., Joseph, J., Abi-Jaoudeh, N., Chang, P. D., Yu, W., & Chow, D. S. (2023). Impact of an automated large vessel occlusion detection tool on clinical workflow and patient outcomes. Frontiers in Neurology, 14. https://doi.org/10.3389/fneur.2023.1179250
U.S. Department of Health and Human Services. How many people are affected by/at risk for stroke?. Eunice Kennedy Shriver National Institute of Child Health and Human Development. https://www.nichd.nih.gov/health/topics/stroke/conditioninfo/risk#:~:text=Each%20year%2C%20about%20795%2C000%20people,another%20stroke%20within%205%20years.
Yu, J., Park, S., Kwon, S.-H., Ho, C. M., Pyo, C.-S., & Lee, H. (2020). AI-based stroke disease prediction system using real-time electromyography signals. Applied Sciences, 10(19), 6791. https://doi.org/10.3390/app10196791