Comparison of single-shot echo-planar imaging with deep learning reconstruction (ss-EPI with DLR) and multi-shot echo-planar imaging with simultaneous multislice (ms-EPI with SMS) for breast DWI: enhanced image quality and reader preference

Deep learning reconstruction (DLR) in single-shot echo-planar imaging (ss-EPI) enhances image quality in diffusion-weighted imaging (DWI) of breast MRI. We compared the image quality and diagnostic performance of ss-EPI with DLR versus multi-shot EPI with simultaneous multi-slice (ms-EPI with SMS). This retrospective study included 77 breast cancer cases from 74 consecutive women (mean age: 53.1 years) who underwent breast DWI. Two radiologists assessed image quality and reader preference, and the cancer detection rate (CDR) was calculated. Quantitative analysis included apparent diffusion coefficient (ADC), signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and lesion contrast, along with phantom measurements. Qualitative analysis showed ss-EPI with DLR had significantly higher scores for fat suppression and overall image quality, while artifact presence and lesion conspicuity were comparable to ms-EPI with SMS. No significant difference in CDR was observed between the two sequences. Both radiologists preferred ss-EPI with DLR (Reader 1: 78.4%; Reader 2: 89.2%). Quantitatively, ss-EPI with DLR had significantly higher lesion contrast for both readers ( p ≤ 0.001), while ADC and SNR values were comparable. Phantom measurements showed no significant differences in ADC or SNR between ss-EPI with DLR and ms-EPI with SMS, although SNR was higher with ss-EPI with DLR (630.03 ± 759.26 vs. SMS: 607.45 ± 565.21, p = 0.912). The acquisition time was shorter for ss-EPI with DLR (2.06 min) than for ms-EPI with SMS (3.29 min). ss-EPI with DLR yielded superior image quality, greater reader preference, and a 40% reduction in scan time compared with ms-EPI with SMS.