Enhancing fMRI quality control

• Published in: Journal of neuroscience methods Vol. 415; p. 110337
• Main Authors: van den Berg, Lennard, Ramsey, Nick, Raemaekers, Mathijs
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier B.V 01.03.2025
• Subjects: Adolescent; Adult; Age Factors; Aged; Brain - diagnostic imaging; Brain - physiology; Brain Mapping - methods; Brain Mapping - standards; Databases, Factual; Female; FMRI; FMRI statistics; Humans; Image Processing, Computer-Assisted - methods; Magnetic Resonance Imaging - methods; Magnetic Resonance Imaging - standards; Male; Memory, Short-Term - physiology; Middle Aged; Quality Control; Sample Size; Task fMRI; Template matching analysis; Young Adult; FMRI statistics; Task fMRI; Template matching analysis; FMRI
• ISSN: 0165-0270; 1872-678X; 1872-678X
• DOI: 10.1016/j.jneumeth.2024.110337

fMRI in clinical settings faces challenges affecting activity maps. Template matching can screen for abnormal results by providing an objective metric of activity map quality. This research tests how sample size, age, or gender-specific templates, and unilateral templates affect template matching results. We used an fMRI database of 76 healthy subjects performing 7 tasks assessing motor, language, and working memory functions. Templates were created with varying numbers of subjects, genders, and ages. Individual subjects were compared to templates using leave-one-out cross validation. We also compared unilateral and bilateral templates. Increasing sample size improved template matches, with diminishing returns for larger sample sizes. Gender and age-specific templates increased correlations for some tasks, with age having a larger effect than gender. Generally, templates including all subjects provided the highest correlations, indicating that age and gender effects did not outweigh the benefits of larger sample sizes. Unilateral templates of the task-dominant hemisphere increased template correlations. Age and gender affect templates, but the benefits depend on the database size. When the database is large enough, age and gender effects are beneficial. Unilateral templates enhance template matching, but practical benefits depend on the severity of neurological abnormalities in patients. •Larger sample sizes when creating templates increases template matching scores.•Using age or gender specific templates can increase template matching scores.•Using age specific templates has larger effects than using gender specific templates.•Non-specific templates can still perform better by being based on larger sample sizes.•Unilateral templates of the task-dominant hemisphere enhances template matching.