Annually, stroke affects millions of people in the USA. Timing is critical in strokes and patient care. Thus, early stroke diagnoses help patients to obtain optimal treatments and interventions, leading to better prognoses. Earlier stroke diagnosis aided by imaging and artificial intelligence is critical in impacting the treatments and outcomes of stroke patients.

Machine learning application using imaging helps to increase the detection accuracy and prognosis of interventions in stroke patients. In this research investigation, an artificial intelligence model is developed to aid in the diagnosis and prognosis of stroke patients and interventions.  

The model was created with machine learning algorithms using over 3,700 clinical patient data from medical institutions. They were applied to develop, train, and test the machine learning model. For the training phase, 60% of the patient data were selected using randomization while the randomized remaining 40% of the patient data was applied for the testing phase in stroke diagnosis and interventional prognosis.  

The artificial intelligence model was able to reach an overall 88.3% accuracy in diagnosis and prognosis of stroke patients and interventional outcomes using clinical patient data and their imaging.

Thus, machine learning can be applied to aid medical professionals and global regions that are underserved in order to enhance early detection, diagnostic accuracy, and outcome prognosis for stroke patients.