Geometry and Morphometry of Gyri and Sulci in Young Adults: Novel Geometric Fingerprints
Poster No:
1722
Submission Type:
Abstract Submission
Authors:
liang he1, Paule Toussaint1, Jiadong Yan1, Rui Ding1, Alejandro Salinas-Medina1, Guan Zhou1, Xiaobo Liu1, Xue Liu1, Alan Evans1
Institutions:
1McGill University, Montreal, Quebec
First Author:
Co-Author(s):
Introduction:
Convex gyri and concave sulci are the prominent features of the convoluted cerebral cortex. They derive from fiber tensor constraints and are influenced by genetic/environmental factors related to cortical function. The extraction of gyro-sulcal (GS) lines and curvature changes has emerged as an effective proxy to simplify the geometric representation of a highly convoluted gyrus/sulcus. However, this is not sufficient to quantify surface mesh patterns and investigate their relationship with cognition. Using the TRACE (Lyu et al., 2018) tool to robustly and reproducibly extract GS-lines, we define a heterogeneous graph to represent each region-of-interest (ROI) and extract geometric features that capture both topological and linear information for each ROI. By distinguishing multiscale cortical structures, GS-lines identify cortical ROIs while providing additional mesh information. This enhanced representation captures morphologic structures and enables the recognition of textural fingerprints and the identification of potential biomarkers.
Methods:
Dataset: We utilized the HCP (Human Connectome Project) S1200 3T dataset. Subjects flagged with Issue Code B (indicating segmentation and/or surface errors) were excluded, along with three additional subjects for whom local gyrification index (LGI) measures could not be calculated. This yielded a final sample of 937 subjects with structural (T1 and T2w) data, diffusion images, and cognitive measurements.
Gyral/Sulcal Geometric and Morphometric Features: We utilized the Destrieux anatomical atlas to obtain 148 gyral and sulcal labels. For topology, we defined three-layer features from the heterogeneous graph representation of each ROI, enabling a comprehensive characterization. We employed 10 linearity measures (Rosin et al., 2016; Stojmenović et al., 2008) for line trend, and 10 common morphometric features called MSN (Morphometric Similarity Network) features, for measuring ROIs.
Model Training for Classification: We employed a multilayer perceptron (MLP) classifier to investigate the geometric characteristics of gyral and sulcal regions. We applied an early-stopping strategy to prevent overfitting during the training process. Prior to training, a 5-fold cross-validation approach was implemented, dividing families into training and test sets.
Gyral/Sulcal Geometric and Morphometric Features: We utilized the Destrieux anatomical atlas to obtain 148 gyral and sulcal labels. For topology, we defined three-layer features from the heterogeneous graph representation of each ROI, enabling a comprehensive characterization. We employed 10 linearity measures (Rosin et al., 2016; Stojmenović et al., 2008) for line trend, and 10 common morphometric features called MSN (Morphometric Similarity Network) features, for measuring ROIs.
Model Training for Classification: We employed a multilayer perceptron (MLP) classifier to investigate the geometric characteristics of gyral and sulcal regions. We applied an early-stopping strategy to prevent overfitting during the training process. Prior to training, a 5-fold cross-validation approach was implemented, dividing families into training and test sets.
Results:
We classified cortical tissues based on topological and linear features of the GS-lines to capture surface mesh organization via an MLP (see Table 1). While typical features derived from multimodal imaging modalities (T1/T2-weighted and diffusion imaging) outperformed mesh features in classification performance, gyral/sulcal features demonstrated strong potential for distinguishing the mesh organization. Importantly, these mesh features are significantly easier to obtain compared to multimodal features and offer additional insights into the cortical structure, as evidenced by the improved classification performance when combining GS and MSN features (see Table 1).
Conclusions:
1. Gyral-sulcal geometric features provide supplementary information for investigating the anatomy and morphology of the brain surface.
2. The gyro-sulcal geometric features are efficient for recognizing gyral and sulcal structure and identifying the specific ROI, thus potentially being the biomarker and geometric fingerprint.
3. These gyro-sulcal geometric features are only based on surface mesh, which is easy to obtain and calculate.
2. The gyro-sulcal geometric features are efficient for recognizing gyral and sulcal structure and identifying the specific ROI, thus potentially being the biomarker and geometric fingerprint.
3. These gyro-sulcal geometric features are only based on surface mesh, which is easy to obtain and calculate.
Modeling and Analysis Methods:
Classification and Predictive Modeling 2
Other Methods
Neuroanatomy, Physiology, Metabolism and Neurotransmission:
Cortical Anatomy and Brain Mapping 1
Keywords:
ADULTS
Cortex
Data analysis
Machine Learning
Morphometrics
STRUCTURAL MRI
Structures
1|2Indicates the priority used for review
Abstract Information
By submitting your proposal, you grant permission for the Organization for Human Brain Mapping (OHBM) to distribute your work in any format, including video, audio print and electronic text through OHBM OnDemand, social media channels, the OHBM website, or other electronic publications and media.
The Open Science Special Interest Group (OSSIG) is introducing a reproducibility challenge for OHBM 2025. This new initiative aims to enhance the reproducibility of scientific results and foster collaborations between labs. Teams will consist of a “source” party and a “reproducing” party, and will be evaluated on the success of their replication, the openness of the source work, and additional deliverables. Click here for more information. Propose your OHBM abstract(s) as source work for future OHBM meetings by selecting one of the following options:
Please indicate below if your study was a "resting state" or "task-activation” study.
Healthy subjects only or patients (note that patient studies may also involve healthy subjects):
Was this research conducted in the United States?
Were any human subjects research approved by the relevant Institutional Review Board or ethics panel? NOTE: Any human subjects studies without IRB approval will be automatically rejected.
Were any animal research approved by the relevant IACUC or other animal research panel? NOTE: Any animal studies without IACUC approval will be automatically rejected.
Please indicate which methods were used in your research:
For human MRI, what field strength scanner do you use?
Which processing packages did you use for your study?
Provide references using APA citation style.
Rosin, P. L., Pantović, J., & Žunić, J. (2016). Measuring linearity of curves in 2D and 3D. Pattern Recognition, 49, 65–78. https://doi.org/10.1016/j.patcog.2015.07.011
Stojmenović, M., Nayak, A., & Zunic, J. (2008). Measuring linearity of planar point sets. Pattern Recognition, 41(8), 2503–2511. https://doi.org/10.1016/j.patcog.2008.01.013