ENHANCING GROUP-WISE CONSISTENCY IN 3-HINGE GYRUS MATCHING VIA ANATOMICAL EMBEDDING AND STRUCTURAL CONNECTIVITY...

Area of Science:

• Neuroscience
• Brain Imaging
• Computational Anatomy

Background:

• A novel cortical folding pattern, the 3-hinge gyrus (3HG), has been identified.
• 3HGs are defined by the convergence of gyri from three directions at crests.
• Existing methods for 3HG analysis lack cross-subject correspondence due to individual anatomical variations.

Purpose of the Study:

• To develop a method for establishing precise one-to-one correspondences for 3-hinge gyri across individuals.
• To leverage multimodal imaging data to improve the accuracy of 3HG identification and comparison.
• To address the limitations of purely anatomical feature-based methods for 3HG analysis.

Main Methods:

• Developed a learning-based method to encode 3HG anatomical features into comparable embedding vectors.
• Integrated structural connectivity patterns from Diffusion Tensor Imaging (DTI) with T1 MRI data.
• Utilized multimodal imaging to overcome the one-to-many matching problem in 3HG identification.

Main Results:

• The multimodal approach significantly improved the accuracy of 3HG correspondences compared to anatomical features alone.
• Effectively mitigated the issue of multiple potential matches for 3HGs based solely on anatomy.
• Demonstrated enhanced precision in identifying one-to-one 3HG matches across subjects.

Conclusions:

• Leveraging multimodal imaging data (T1 MRI and DTI) enhances the precision of 3-hinge gyrus correspondence.
• This approach improves the reliability of 3HG identification for group-level brain analyses.
• The findings contribute to the broader utilization of 3HGs in neuroscience research.