Deep Learning-Enhanced T1-Weighted Imaging for Breast MRI at 1.5T
<b>Background/Objectives</b>: Assessment of a novel deep-learning (DL)-based T1w volumetric interpolated breath-hold (VIBE<sub>DL</sub>) sequence in breast MRI in comparison with standard VIBE (VIBE<sub>Std</sub>) for image quality evaluation. <b>Methods<...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-07-01
|
| Series: | Diagnostics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2075-4418/15/13/1681 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | <b>Background/Objectives</b>: Assessment of a novel deep-learning (DL)-based T1w volumetric interpolated breath-hold (VIBE<sub>DL</sub>) sequence in breast MRI in comparison with standard VIBE (VIBE<sub>Std</sub>) for image quality evaluation. <b>Methods</b>: Prospective study of 52 breast cancer patients examined at 1.5T breast MRI with T1w VIBE<sub>Std</sub> and T1 VIBE<sub>DL</sub> sequence. T1w VIBE<sub>DL</sub> was integrated as an additional early non-contrast and a delayed post-contrast scan. Two radiologists independently scored T1w VIBE <sub>Std/DL</sub> sequences both pre- and post-contrast and their calculated subtractions (SUBs) for image quality, sharpness, (motion)–artifacts, perceived signal-to-noise and diagnostic confidence with a Likert-scale from 1: Non-diagnostic to 5: Excellent. Lesion diameter was evaluated on the SUB for T1w VIBE<sub>Std</sub>/<sub>DL</sub>. All lesions were visually evaluated in T1w VIBE<sub>Std</sub>/<sub>DL</sub> pre- and post-contrast and their subtractions. Statistics included correlation analyses and paired t-tests. <b>Results</b>: Significantly higher Likert scale values were detected in the pre-contrast T1w VIBE<sub>DL</sub> compared to the T1w VIBE<sub>Std</sub> for image quality (each <i>p</i> < 0.001), image sharpness (<i>p</i> < 0.001), SNR (<i>p</i> < 0.001), and diagnostic confidence (<i>p</i> < 0.010). Significantly higher values for image quality (<i>p</i> < 0.001 in each case), image sharpness (<i>p</i> < 0.001), SNR (<i>p</i> < 0.001), and artifacts (<i>p</i> < 0.001) were detected in the post-contrast T1w VIBE<sub>DL</sub> and in the SUB. SUB<sub>DL</sub> provided superior diagnostic certainty compared to SUB<sub>Std</sub> in one reader (<i>p</i> = 0.083 or <i>p</i> = 0.004). <b>Conclusions</b>: Deep learning-enhanced T1w VIBE<sub>DL</sub> at 1.5T breast MRI offers superior image quality compared to T1w VIBE<sub>Std</sub>. |
|---|---|
| ISSN: | 2075-4418 |