Annually, millions of lives are affected and impacted by strokes. Timing plays a critical role in the management and treatment of strokes in patients. Thus, rapid, accurate, and early detection and diagnosis of stroke greatly increases chances of survival and overall prognosis. To enhance early detection and diagnosis of stroke, artificial intelligence such as machine learning and biomedical imaging can be utilized to optimize the treatment outcomes of patients.  

Machine learning applications using imaging aid early detection and diagnostic accuracy, improving management and interventional prognosis in patients with strokes. In this research, artificial intelligence using machine learning algorithms is applied to enhance detection and diagnosis of strokes in patients.   

The model was created with machine learning algorithms using over 2,368 patient data from medical facilities, clinics, and hospitals. They were applied for building, training, and testing the artificial intelligence model. To train it, random selection of 50% of the patient data were applied; to test it, the remaining 50% of the patient data was applied for testing its detection of strokes and diagnosis capabilities.  

The machine learning model detected strokes with an overall 90.1% diagnostic accuracy using imaging and the clinical patient data, increasing efficiency and early accurate detection of strokes in patients.

In conclusion, machine learning can be used to aid medical professionals and especially resource-limited communities with early detection and accurate diagnosis of strokes in patients, enhancing outcomes.