Quantitative double echo steady state T2 mapping of upper extremity peripheral nerves and muscles

Area of Science:

• Medical Imaging
• Neurology
• Artificial Intelligence in Medicine

Background:

• T2 mapping is crucial for characterizing peripheral neuropathy and muscle denervation.
• Three-dimensional double echo steady-state (DESS) provides simultaneous 3D qualitative information and T2 maps.
• Insufficient signal-to-noise ratio in DESS can bias T2 values, impacting quantitative accuracy.

Purpose of the Study:

• To evaluate the impact of deep learning reconstruction (DLR) on DESS-T2 values.
• To assess the feasibility of using DESS-T2 maps for differentiating abnormal from normal nerves and muscles in the upper extremities, using electromyography as a reference standard.

Main Methods:

• Analysis of imaging data from 25 subjects.
• Comparison of DESS-T2 values with and without DLR in muscles and nerves.
• Correlation of DESS-T2 values with electromyography findings for muscle abnormalities.

Main Results:

• DLR significantly decreased DESS-T2 values in both abnormal and normal muscles, indicating noise reduction.
• Mean DESS-T2 values were significantly higher in abnormal nerves and muscles compared to normal ones, irrespective of DLR.
• Elevated muscle DESS-T2 values correlated with electromyography-determined motor unit recruitment.

Conclusions:

• Deep learning reconstruction enhances the quantitative accuracy of DESS-T2 values.
• Quantitative DESS-T2 mapping effectively differentiates nerves and muscles affected by peripheral neuropathy from healthy tissues.