Dataset Information

Age estimation using cortical surface pattern combining thickness with curvatures.

ABSTRACT: Brain development and healthy aging have been proved to follow a specific pattern, which, in turn, can be applied to help doctors diagnose mental diseases. In this paper, we design a cortical surface pattern (CSP) combining the cortical thickness with curvatures, which constructs an accurate human age estimation model with relevance vector regression. We test our model with two public databases. One is the IXI database (360 healthy subjects aging from 20 to 82 years old were selected), and the other is the INDI database (303 subjects aging from 7 to 22 years old were selected). The results show that our model can achieve as small as 4.57 years deviation in the IXI database and 1.38 years deviation in the INDI database. Furthermore, we employ this surface pattern to age groups classification and get a remarkably high accuracy (97.77 %) and a significantly high sensitivity/specificity (97.30/98.10 %). These results suggest that our designed CSP combining thickness with curvatures is stable and sensitive to brain development, and it is much more powerful than voxel-based morphometry used in previous methods for age estimation.

SUBMITTER: Wang J

PROVIDER: S-EPMC3961503 | biostudies-literature | 2014 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Age estimation using cortical surface pattern combining thickness with curvatures.

Wang Jieqiong J Li Wenjing W Miao Wen W Dai Dai D Hua Jing J He Huiguang H

Medical & biological engineering & computing 20140107 4

Brain development and healthy aging have been proved to follow a specific pattern, which, in turn, can be applied to help doctors diagnose mental diseases. In this paper, we design a cortical surface pattern (CSP) combining the cortical thickness with curvatures, which constructs an accurate human age estimation model with relevance vector regression. We test our model with two public databases. One is the IXI database (360 healthy subjects aging from 20 to 82 years old were selected), and the o ...[more]

PMID: 24395657

Similar Datasets

Project description:Importance:Magnetic resonance imaging (MRI) studies of aging adults have shown substantial intersubject variability across various brain metrics, and some of this variability is likely attributable to chronological age being an imprecise measure of age-related change. Accurately quantifying one's biological age could allow better quantification of healthy and pathological changes in the aging brain. Objective:To investigate the association of DNA methylation (DNAm)-based biological age with cortical thickness and to assess whether biological age acceleration compared with chronological age captures unique variance in cortical thinning. Design, Setting, and Participants:This cross-sectional study used high-resolution structural brain MRI data collected from a sample of healthy aging adults who were participating in a larger ongoing neuroimaging study that began in May 2014. This population-based study accrued participants from the greater Omaha, Nebraska, metropolitan area. One hundred sixty healthy adults were contacted for the MRI component, 82 of whom participated in both DNAm and MRI study components. Data analysis was performed from March to June 2019. Main Outcomes and Measures:Vertexwise cortical thickness, DNAm-based biological age, and biological age acceleration compared with chronological age were measured. A pair of multivariable regression models were computed in which cortical thickness was regressed on DNAm-based biological age, controlling for sex in the first model and also controlling for chronological age in the second model. Results:Seventy-nine adult participants (38 women; mean [SD] age,?43.82 [14.50] years; age range, 22-72 years) were included in all final analyses. Advancing biological age was correlated with cortical thinning across frontal, superior temporal, inferior parietal, and medial occipital regions. In addition, biological age acceleration relative to chronological age was associated with cortical thinning in orbitofrontal, superior and inferior temporal, somatosensory, parahippocampal, and fusiform regions. Specifically, for every 1 year of biological age acceleration, cortical thickness would be expected to decrease by 0.024 mm (95% CI, -0.04 to -0.01 mm) in the left orbitofrontal cortex (partial r, -0.34; P?=?.002), 0.014 mm (95% CI, -0.02 to -0.01 mm) in the left superior temporal gyrus (partial r, -0.36; P?=?.001), 0.015 mm (95% CI, -0.02 to -0.01 mm) in the left fusiform gyrus (partial r, -0.38; P?=?.001), 0.015 mm (95% CI, -0.02 to -0.01 mm) in the right fusiform gyrus (partial r, -0.43; P?<?.001), 0.019 mm (95% CI, -0.03 to -0.01 mm) in the right inferior temporal sulcus (partial r, -0.34; P?=?.002), and 0.011 mm (95% CI, -0.02 to -0.01 mm) in the right primary somatosensory cortex (partial r, -0.37; P?=?.001). Conclusions and Relevance:To our knowledge, this is the first study to investigate vertexwise cortical thickness in relation to DNAm-based biological age, and the findings suggest that this metric of biological age may yield additional insight on healthy and pathological cortical aging compared with standard measures of chronological age alone.

Dataset Information

Age estimation using cortical surface pattern combining thickness with curvatures.

Publications

Age estimation using cortical surface pattern combining thickness with curvatures.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets