SEGMENTATION-FREE MEASURING OF CORTICAL THICKNESS FROM MRI

Area of Science:

• Neuroimaging
• Medical Image Analysis
• Computational Neuroscience

Background:

• Cortical thickness estimation is crucial in Magnetic Resonance (MR) brain imaging.
• Existing methods often rely on pre-segmentation, which can lose information due to noise and partial volume effects.

Purpose of the Study:

• To present a novel framework for measuring cortical thickness in MR brain imaging.
• To improve upon traditional methods by leveraging voxel-wise gray matter probability maps.

Main Methods:

• The proposed method minimizes line integrals over the probability map of gray matter in the MRI volume.
• This approach avoids hard classification, preserving information from underlying class probabilities.

Main Results:

• The framework's performance was validated using both artificial datasets and real MR imaging data.
• Evaluations included data from both healthy subjects and those affected by Alzheimer's disease.

Conclusions:

• The novel method offers a more robust approach to cortical thickness measurement in MR brain imaging.
• This technique effectively utilizes probabilistic information, potentially enhancing diagnostic capabilities for neurological conditions.