A comprehensive and reliable protocol for manual segmentation of the human claustrum using high-resolution MRI

• Published in: Brain Structure and Function Vol. 230; no. 7; p. 134
• Main Authors: Kang, Steven Seung-Suk, Bodenheimer, Joseph, Morris, Kayley, Butler, Tracy
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 13.08.2025; Springer Nature B.V
• Subjects: Adult; Basal ganglia; Biomedical and Life Sciences; Biomedicine; Brain; Cell Biology; Claustrum - anatomy & histology; Claustrum - diagnostic imaging; Convulsions & seizures; Female; Human subjects; Humans; Image Processing, Computer-Assisted - methods; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Male; Medical imaging; Mental disorders; Morphology; Network switching; Neural networks; Neuroimaging; Neuroimaging - methods; Neurology; Neurosciences; Original Article; Reproducibility of Results; Segmentation; Substantia grisea; Young Adult; Young adults; Manual segmentation protocol; Subregions; Reliability; MRI; The human claustrum
• ISSN: 1863-2661; 1863-2653; 1863-2661; 0340-2061
• DOI: 10.1007/s00429-025-02993-7

The claustrum is a thin gray matter structure in each brain hemisphere, characterized by exceptionally high connectivity with nearly all brain regions. Despite extensive animal studies on its anatomy and function and growing evidence of claustral deficits in neuropsychiatric disorders, its specific roles in normal and abnormal human brain function remain largely unknown. This is primarily due to its thin and complex morphology, which limits accurate anatomical delineation and neural activity isolation in conventional in vivo neuroimaging. To facilitate future neuroimaging studies, we developed a comprehensive and reliable manual segmentation protocol based on a cellular-resolution brain atlas and high-resolution (0.7 mm isotropic) MRI data. The protocols involve detailed guidelines to delineate the entire claustrum, including the inferior parts that have not been clearly described in earlier MRI studies. Additionally, we propose a geometric method to parcellate the claustrum into three subregions (the dorsal, ventral, and temporal claustrum) along the superior-to-inferior axis. The mean bilateral claustrum volume in 10 young adults was 3307.5 mm 3 , approximately 0.21% of total intracranial volume. Our segmentation protocol demonstrated high inter- and intra-rater reliability (ICC > 0.89, DSC > 0.85), confirming its replicability. This comprehensive and reliable manual segmentation protocol offers a robust foundation for anatomically precise neuroimaging investigations of the human claustrum.