The role of diversity in data‐driven analysis of multi‐subject fMRI data: Comparison of approaches based on independence and sparsity using global performance metrics

• Published in: Human brain mapping Vol. 40; no. 2; pp. 489 - 504
• Main Authors: Long, Qunfang, Bhinge, Suchita, Levin‐Schwartz, Yuri, Boukouvalas, Zois, Calhoun, Vince D., Adalı, Tülay
• Format: Journal Article
• Language: English
• Published: Hoboken, USA John Wiley & Sons, Inc 01.02.2019
• Subjects: Adult; Algorithms; Brain - diagnostic imaging; Brain - physiology; Brain - physiopathology; Data analysis; Data Interpretation, Statistical; Data processing; data‐driven analysis; dictionary learning; diversity; Female; fMRI analysis; Functional magnetic resonance imaging; Functional Neuroimaging - methods; Functional Neuroimaging - standards; global metric; Humans; ICA; Image Processing, Computer-Assisted - methods; Image Processing, Computer-Assisted - standards; independence; Independent component analysis; Magnetic Resonance Imaging; Male; Mental disorders; Middle Aged; Nerve Net - diagnostic imaging; Nerve Net - physiology; Nerve Net - physiopathology; performance evaluation; Performance measurement; Principal Component Analysis; Schizophrenia; Schizophrenia - diagnostic imaging; Schizophrenia - physiopathology; Sparsity; Statistics; diversity; ICA; data-driven analysis; global metric; performance evaluation; sparsity; dictionary learning; independence; fMRI analysis

Data‐driven methods have been widely used in functional magnetic resonance imaging (fMRI) data analysis. They extract latent factors, generally, through the use of a simple generative model. Independent component analysis (ICA) and dictionary learning (DL) are two popular data‐driven methods that ar...