Abstract:Rationale and objective: A single-shot T2-weighted deep-learning-based image reconstruction (DL-HASTE) has been recently developed allowing for shorter acquisition time than conventional half-Fourier acquisition single-shot turbo-spin echo (HASTE). The purpose of this study was to compare image quality of conventional 6 mm HASTE with DL-HASTE at 4 mm and 6 mm slice thickness. Materials and methods: 91 patients (51 female; mean±SD age: 44±10years) who underwent 3T MR enterography from 5/15/2023-7/15/2023 including pelvic conventional HASTE and DL-HASTE were included. Patients either had 4 mm-DL-HASTE or 6 mm-DL-HASTE. Four abdominal radiologists, blinded to sequence type, independently evaluated overall image quality, artifacts over bowel, bowel wall sharpness, and confidence for the presence/absence of bowel abnormalities on 5-point Likert scales. Readers recorded the presence/absence of ileal wall thickening, ileal inflammation, stricture, and penetrating disease on each sequence. Wilcoxon signed-rank test with continuity correction was used for paired comparisons and Wilcoxon rank sum test was used for unpaired ordinal comparisons. A p < .05 indicated statistical significance. Results: Acquisition times for 6 mm HASTE, 4 mm-DL-HASTE, and 6 mm-DL-HASTE were 64 s, 51 s, and 49 s, respectively. Overall image quality and bowel sharpness were significantly improved for 4 mm-DL-HASTE versus HASTE for 3/4 readers (all p < .05) and similar for the 4th reader (p > .05). Diagnostic confidence was similar for all readers (p > .05). 6 mm-DL-HASTE was similar to HASTE for bowel sharpness, image quality, and confidence for 3/4 readers (all p > .05). The presence of ileal thickening, ileal inflammation, stricture, and penetrating disease were similar for all readers for HASTE, 4 mm-DL-HASTE, and 6 mm-DL-HASTE (all p > .05). Conclusion: 4 mm-DL-HASTE had superior image quality than conventional HASTE at shorter acquisition time.

Accelerated T2-weighted MRI of the bowel at 3T using a single-shot technique with deep learning-based image reconstruction: impact on image quality and disease detection