Anatomy-guided Dense Individualized and Common Connectivity-based Cortical Landmarks (A-DICCCOL)

Area of Science:

• Neuroscience
• Brain Mapping
• Computational Anatomy

Background:

• Establishing reproducible human brain structural and functional correspondences is crucial for brain mapping.
• Previous Dense Individualized and Common Connectivity-based Cortical Landmarks (DICCCOL) identified 358 reproducible connectional landmarks using DTI data.
• Integrating anatomical and morphological data can improve landmark initialization and optimization.

Purpose of the Study:

• To present a novel anatomy-guided landmark discovery framework (A-DICCCOL).
• To define and optimize brain landmarks by integrating anatomical, morphological, and fiber connectional information.
• To enhance the reproducibility and accuracy of brain mapping across individuals and populations.

Main Methods:

• Developed an energy minimization framework integrating anatomical, morphological, and fiber connectional data.
• Used the framework for landmark initialization, group-wise optimization, and prediction.
• Validated the discovered landmarks for reproducibility, predictability, and anatomical/functional correspondences.

Main Results:

• Identified 555 consistent connectional landmarks using the anatomy-guided framework.
• The new landmarks, termed A-DICCCOL, demonstrated reproducibility and predictability across individuals and populations.
• A-DICCCOL landmarks showed accurate anatomical, connectional, and functional correspondences.

Conclusions:

• The A-DICCCOL system successfully integrates diverse neuroimaging data for robust landmark identification.
• This system enhances common cortical architecture representation with anatomical, connectional, and functional correspondences.
• A-DICCCOL offers significant potential for diverse applications in brain science and personalized medicine.