Performance of fast learning approach to predicting black fungus diseases

Abstract

Microscopic examination can be used to make a preliminary diagnosis of fungal infections. Due to their apparent similarities, it frequently does not allow for the species to be identified clearly. Therefore, additional biochemical tests are typically required. This adds extra expenses and can make the identification process last up to ten days. Given the high death rate for immunosuppressed patients, such a delay in the adoption of targeted therapy could have serious consequences. The fast learning network method is an alternative that provides information with a unique approach to predicting black fungus. The experimental results of prediction showed that the performance of the fast learning method is superior as compared with the five algorithms used in this paper.