医疗卫生装备2025，Vol.46Issue(3)：59-63,5.DOI:10.19745/j.1003-8868.2025049
基于深度学习的HASTE序列在胰腺T2WI中的应用价值
Application value of HASTE sequence based on deep learning in pancreatic T2WI
曹达 1王传兵 1史红媛 1郝绍伟 2李青1
作者信息
- 1. 南京医科大学第一附属医院放射科,南京 210029
- 2. 西门子数字医疗科技(上海)有限公司应用培训部,上海 200131
- 折叠
摘要
Abstract
Objective To evaluate the clinical application value of the half-Fourier acquisition single-shot turbo spin echo(HASTE)sequence based on deep learning(DL)in pancreatic T2WI.Methods Data were collected from 41 patients who un-derwent both BLADE and DL-HASTE sequence scans during pancreatic T2WI at some hospital from February to July 2023.Qualitative assessments were made regarding overall image quality,pancreatic edge sharpness,pancreatic duct edge sharp-ness,pancreatic duct visua-lization(proximal,middle and distal segments)and lesion visibility of BLADE-sequence and DL-HASTE-sequence images.Quantitative assessments were carried out in terms of scan time,signal-to-noise ratio(SNR)and contrast-to-noise ratio(CNR).Statistical analyses were performed using SPSS 24.0.Results DL-HASTE sequence behaved significantly better than BLADE in pancreatic duct edge sharpness,pancreatic duct visualization,lesion visibility,scan time and CNR,while worse in pane-reatic edge sharpness(all P＜0.05).There were significant differences between DL-HASTE and BLADE sequences in overall image quality and SNR(all P＞0.05).Conclusion The DL-HASTE sequence maintains image quality while significantly shor-tening scan time,making it suitable for patients with irregular respiratory rates.[Chinese Medical Equipment Journal,2025,46(3):59-63]关键词
磁共振成像/T2WI/深度学习/脂肪抑制序列/HASTE序列/胰腺疾病Key words
magnetic resonance imaging/T2WI/deep learning/fat suppression sequence/half-Fourier acquisition single-shot turbo spin echo sequence/pancreatic disease分类
医药卫生引用本文复制引用
曹达,王传兵,史红媛,郝绍伟,李青..基于深度学习的HASTE序列在胰腺T2WI中的应用价值[J].医疗卫生装备,2025,46(3):59-63,5.
