Application value of T2 fluid-attenuated inversion recovery sequence based on deep learning in static lacunar...

Area of Science:

• Radiology
• Medical Imaging
• Artificial Intelligence

Background:

• Static lacunar infarction (SLI) lesion monitoring is crucial for disease management and prognosis.
• Magnetic resonance imaging (MRI) is a key tool for SLI lesion surveillance.

Purpose of the Study:

• To evaluate the image quality of T2 fluid-attenuated inversion recovery (T2-FLAIR) sequences using artificial intelligence-assisted compressed sensing (ACS).
• To assess the clinical applicability of ACS in detecting SLI lesions.

Main Methods:

• 42 patients prospectively underwent T2-FLAIR MRI scans.
• Two readers assessed image quality (overall and lesion-specific) and lesion detection using 1D acceleration and ACS (factors 2, 3, 4).
• Quantitative analysis included signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR).

Main Results:

• Subjective image quality assessments were consistent between readers.
• ACS (factors 2, 3, 4) demonstrated significantly higher CNR compared to 1D acceleration.
• No significant difference in SNR or lesion detection was found between 1D, ACS2, and ACS3 modes, but ACS3 reduced scan time by 40%.

Conclusions:

• ACS acceleration mode substantially reduces MRI scan time for T2-FLAIR sequences.
• ACS-accelerated T2-FLAIR imaging provides good SNR, high CNR, and effective SLI lesion detection.
• ACS shows strong clinical applicability for monitoring SLI lesions.