A couple weeks ago, a teacher saved the grandmother of one of her students from a stroke after noticing symptoms over Zoom. In the same vein, a recent Medgadget article discussed a new tool for rapidly diagnosing strokes. The technology, developed by researchers at Penn State and Houston Methodist Hospital, uses a smartphone to record the speech and facial movements of the user. The data is then processed with a machine-learning algorithm to determine whether a stroke likely occurred.
The system has proven as accurate as an ER clinician in diagnosing a stroke (~79%), and it does so within just minutes. Because millions of neurons die each minute during a stroke, rapid diagnosis and treatment is imperative to limit long-term damage. Accuracy of the test will likely improve as the dataset grows.
