Improving classifier performance in neuroimaging data with the help of optimal machine learning technique

Abstract

In the past decades, the study of human brain evolved rapidly through radiographic imaging techniques in both healthy and diseased state. Neuroimaging considered as a segment of brain image mapping that exhibit a peerless form of technique. Medical-imaging tools like magnetic resonance (MR) scanners, is most frequently used sources of quantitively data on brain
structure. Study of these data results in complex data involving multidimensional images to noisy data, this leads to misinterpretation. Even with its analyses, the diagnose predictions of clinical assets has not been preferred to neuroimaging, hence we use machine learning algorithms.