PROJECT TITLE :
Discriminating Bipolar Disorder From Major Depression Based on SVM-FoBa: Efficient Feature Selection With Multimodal Brain Imaging Data
Discriminating between bipolar disorder (BD) and major depressive disorder (MDD) is a major clinical challenge thanks to the absence of known biomarkers; hence a better understanding of their pathophysiology and brain alterations is urgently required. Given the complexity, feature selection is particularly necessary in neuroimaging applications, but, feature dimension and model understanding present serious challenges. During this study, a novel feature selection approach based mostly on linear support vector machine with a forward-backward search strategy (SVM-FoBa) was developed and applied to structural and resting-state practical magnetic resonance imaging information collected from twenty one BD, 25 MDD and twenty three healthy controls. Discriminative options were drawn from each knowledge modalities, with that the classification of BD and MDD achieved an accuracy of 92.one% (one thousand bootstrap resamples). Weight analysis of the selected options more revealed that the inferior frontal gyrus might characterize a central role in BD-MDD differentiation, additionally to the default mode network and also the cerebellum. A modality-wise comparison also prompt that functional information outweighs anatomical by a massive margin when classifying the two clinical disorders. This work validated the advantages of multimodal joint analysis and therefore the effectiveness of SVM-FoBa, that has potential to be used in identifying attainable biomarkers for several mental disorders.
Did you like this research project?
To get this research project Guidelines, Training and Code... Click Here