A rapid compression technique for 4-D functional MRI images using data rearrangement and modified binary array...

Area of Science:

• Medical Imaging
• Data Compression
• Signal Processing

Background:

• Efficient storage and transfer of medical image data are crucial.
• Existing compression techniques for medical images are often time-consuming, hindering their practical application.
• Functional magnetic resonance imaging (fMRI) generates large datasets requiring effective compression solutions.

Purpose of the Study:

• To propose a rapid 4-D lossy compression method for fMRI images.
• To enhance compression efficiency and reduce processing time compared to existing methods.
• To maintain high image quality and diagnostic feature preservation in compressed fMRI data.

Main Methods:

• A novel 4-D lossy compression approach combining data rearrangement, wavelet-based contourlet transform (WBCT), and a modified binary array technique.
• WBCT decomposes wavelet transform sub-bands to capture directional information, followed by a repositioning algorithm to manage coefficient relationships.
• A modified binary array technique compresses quantized coefficients by coding frequent values once.

Main Results:

• The proposed method demonstrated significantly reduced processing time compared to wavelet-based set partitioning in hierarchical trees and SPECK compression schemes.
• Achieved superior compression performance over wavelet-based SPECK coders.
• Maintained a peak signal-to-noise ratio (PSNR) above 70, Structural Similarity Index (SSIM) of 1, and Correlation Coefficient (CC) greater than 0.9.

Conclusions:

• The developed rapid 4-D compression method offers a faster and more efficient solution for fMRI data.
• The technique successfully preserves crucial diagnostic features in reconstructed fMRI images.
• This approach represents a significant advancement in medical image compression technology.