RANDOM FOREST FLAIR RECONSTRUCTION FROM T₁, T₂, AND P -WEIGHTED MRI

Area of Science:

• Medical Imaging
• Neuroscience
• Radiology

Background:

• Fluid Attenuated Inversion Recovery (FLAIR) MRI is crucial for multiple sclerosis (MS) lesion detection.
• MS lesions appear hyperintense on FLAIR, enabling effective segmentation and load calculation.
• Poor image quality or missing FLAIR data can hinder consistent analysis of MS progression.

Purpose of the Study:

• To develop a method for reconstructing FLAIR images from other MRI sequences.
• To address the challenge of missing or artifact-prone FLAIR data in MS studies.
• To provide a reliable surrogate for high-quality FLAIR images in automated MS lesion analysis.

Main Methods:

• Utilized random forest regression for image reconstruction.
• Employed T1-weighted, T2-weighted, and P-weighted MRI images as input.
• Developed a computational approach to generate FLAIR images.

Main Results:

• Reconstructed FLAIR images closely resemble true high-quality FLAIR scans.
• The generated images serve as effective surrogates for tissue segmentation.
• The method successfully addresses data gaps and quality issues in FLAIR imaging.

Conclusions:

• Reconstructed FLAIR images are a viable alternative when true FLAIR data is compromised.
• This technique enhances the consistency and reliability of automated MS lesion analysis.
• The approach facilitates more robust correlation of lesion load with MS disease progression.