A new tool could diagnose a stroke based on abnormalities in a patient’s speech ability and facial muscular movements with the accuracy of an emergency room physician — all within minutes from an interaction with a smartphone. A machine learning model aids in, and potentially speeds up, the diagnostic process by physicians in a clinical setting.
Currently, physicians have to use their past training and experience to determine at what stage a patient should be sent for a CT scan. The new tool analyzes the presence of stroke among actual emergency room patients with suspicion of stroke by using computational facial motion analysis and natural language processing to identify abnormalities in a patient’s face or voice, such as a drooping cheek or slurred speech.
The results could help emergency room physicians to more quickly determine critical next steps for the patient. Ultimately, the application could be utilized by caregivers or patients to make self-assessments before reaching the hospital.
To train the computer model, the researchers built a dataset from more than 80 patients experiencing stroke symptoms at a hospital. Each patient was asked to perform a speech test to analyze their speech and cognitive communication while being recorded on an Apple iPhone.
Testing the model on the hospital dataset, the researchers found that its performance achieved 79 percent accuracy — comparable to clinical diagnostics by emergency room doctors, who use additional tests such as CT scans; however, the model could help save valuable time in diagnosing a stroke, with the ability to assess a patient in as little as four minutes. In severe strokes, it is obvious to doctors from the moment the patient enters the emergency department, but studies suggest that in the majority of strokes with mild symptoms, a diagnosis can be delayed by hours and by then, a patient may not be eligible for the best possible treatments.
Physicians currently use a binary approach toward diagnosing strokes: They either suspect a stroke, sending the patient for a series of scans that could involve radiation or they do not suspect a stroke, potentially overlooking patients who may need further assessment.
For more information, contact Jessica Hallman at