Risk Detection of Stroke Using a Feature Selection and Classification Method

Stroke places a heavy burden of care on global societies. Risk detection of stroke is a challenging and time-sensitive task across the world. This article investigated biomedical tests and electronic archives of 792 records that contained 398 records from the five years preceding the onset of stroke at a community hospital. The records included 28 features. We have proposed a new feature selection model that combines support vector machines with the glow-worm swarm optimization algorithm based on the standard deviation of the features. The results showed that the proposed model achieved 82.58% accuracy by means of the 18 features among the original data set. The new map thus represents an effective detection that can help to identify patients with an increased risk of stroke events.