Brain White Matter Hyperintensity Lesion Characterization in T₂ Fluid-Attenuated Inversion Recovery Magnetic...

Area of Science:

• Neuroimaging
• Biomedical Engineering
• Gerontology

Background:

• Current characterization of age-related white matter hyperintensity (WMH) lesions on T2 fluid-attenuated inversion recovery (FLAIR) magnetic resonance images (MRI) primarily focuses on size and location.
• Systematic morphological characterization of WMH lesions has been lacking, hindering a deeper understanding of their progression.

Purpose of the Study:

• To develop and validate innovative methods for quantifying the shape and texture of WMH lesions.
• To establish a framework for clustering WMH lesions based on morphological and textural features.
• To investigate the potential of these features in predicting lesion growth.

Main Methods:

• Developed a novel method to quantify WMH lesion shape using Zernike transformation and texture using fuzzy logic.
• Employed a multi-dimension feature vector approach for clustering WMH lesions.
• Utilized a region-growing algorithm to calculate the potential growth index (PGI) based on edge intensity distributions.

Main Results:

• Significant differences in PGI were observed among WMH clusters based on shape features (P = 1.06 × 10⁻²).
• Highly significant differences in PGI were found among WMH clusters based on texture features (P < 1 × 10⁻²⁰).
• The developed methods successfully characterized and clustered WMH lesions from T2 FLAIR MRI scans of cognitively normal older adults.

Conclusions:

• Proposed a systematic framework for quantifying shape and texture features of WMH lesions.
• Demonstrated the potential of these quantified features to predict lesion growth in older adults.
• Highlighted the importance of morphological and textural analysis beyond size and location for understanding WMH progression.