A Univariate Morphometry Index for AD-induced Abnormal Cortical Surface Pattern Similarity Measurement

• Published in: 2022 4th International Conference on Industrial Artificial Intelligence (IAI) pp. 1 - 6
• Main Authors: Qu, Zongshuai, Li, Nan, Wang, Gang
• Format: Conference Proceeding
• Language: English
• Published: IEEE 24.08.2022
• Subjects: Atrophy; Cerebral cortex; Magnetic resonance imaging; Morphology; Neuroimaging; Surface morphology; Volume measurement
• DOI: 10.1109/IAI55780.2022.9976557

Impairments in cognition induced by Alzheimer's disease (AD) are closely related to the changes in cerebral cortex morphology. In order to observe AD development and evaluate the effectiveness of interventions in the early stages of AD, we proposed a AD-specific abnormal surface pattern similarity measure capable of exploiting univariate neurodegeneration biomarkers (UNB) to reflect the morphological changes induced by AD-related diseases based on the structural magnetic resonance imaging (sMRI). First, registering the thickness information and anatomical regions of interest (ROIs) between individual cortical surfaces. Second, using the intrinsic thickness information of the \mathbf{A}\boldsymbol{\upbeta}+ AD and the \mathbf{A}\boldsymbol{\upbeta}- normal control (NC) groups via general linear model, we identified the top six ROIs with the most significant surface morphometry changes. Finally, a univariate morphometry index (UMI) of the individual subject was constructed by comparing the similarity between the individual morphological atrophy pattern and the atrophy pattern of the \mathbf{A}\boldsymbol{\upbeta}+ AD group on the identified ROIs. We validated our work in the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort. With the computed UMIs, the estimated minimum sample sizes needed to detect a 25% reduction in the mean annual change with 80% power and two-tailed \mathbf{P}=\boldsymbol{0.05} were 156, 349 and 423 for the longitudinal \mathbf{A}pmb{\upbeta}+\boldsymbol{\text{AD}}, \mathbf{A}\boldsymbol{\upbeta}+ mild cognitive impairment (MCI) and \mathbf{A}\boldsymbol{\upbeta}+ NC groups, respectively. Our experimental results outperformed traditional volume measures and demonstrated that UMI could be used as potential UNB which could reflect the cerebral cortex morphological changes induced by AD-related diseases.