Brendel Jan M, Dehdab Reza, Herrmann Judith, Ursprung Stephan, Werner Sebastian, Almansour Haidara, Weiland Elisabeth, Nickel Dominik, Nikolaou Konstantin, Afat Saif, Gassenmaier Sebastian
Department of Radiology, Diagnostic and Interventional Radiology, Tuebingen University Hospital, University of Tuebingen, Hoppe-Seyler-Str. 3, 72076, Tuebingen, Germany.
MR Application Predevelopment, Siemens Healthineers, Forchheim, Germany.
Radiol Med. 2025 Mar 18. doi: 10.1007/s11547-025-01987-z.
This study aimed to compare a conventional three-dimensional (3-D) magnetic resonance cholangiopancreatography (MRCP) sequence with a deep learning (DL)-accelerated MRCP sequence (hereafter, MRCP) regarding acquisition time and image quality.
We conducted a prospective study of consecutive patients referred for MRCP between November 2023 and April 2024 at a single tertiary center. Each participant underwent 1.5T 3-D T2-weighted turbo spin echo MRCP using both a conventional sequence (threefold acceleration) and MRCP (eightfold acceleration). Three blinded readers independently evaluated image quality, including background signal suppression, bile and pancreatic duct visibility, artifact level, and diagnostic confidence on an ordinal four-point scale. Acquisition times were compared using a paired t-test. Image quality parameters were assessed with repeated measures ANOVA. Interreader agreement was analyzed using Fleiss' κ.
Out of 419 consecutive patients, 30 participants were evaluated (mean age, 63 ± 15 years; 16 men, 14 women). The mean acquisition time was 10:30 ± 03:04 min for conventional MRCP and 3:57 ± 01:13 min for MRCP, P < 0.001. MRCP reduced acquisition time by 62.4%. Artifact levels were rated at 3.17 ± 0.77 for conventional MRCP and 3.56 ± 0.66 for MRCP (P = 0.041). Background signal suppression, bile duct visibility, pancreatic duct visibility, and diagnostic confidence did not differ significantly (P > 0.05). Interreader agreement was substantial to almost perfect (κ: 0.64-87).
Deep learning-accelerated 3-D MRCP reduced acquisition time by 62%, minimized artifacts, and preserved bile and pancreatic duct visibility, supporting its adoption in routine clinical practice.
本研究旨在比较传统三维（3-D）磁共振胰胆管造影（MRCP）序列与深度学习（DL）加速的MRCP序列（以下简称MRCP）在采集时间和图像质量方面的差异。
我们在一家单一的三级中心对2023年11月至2024年4月期间连续转诊进行MRCP检查的患者进行了一项前瞻性研究。每位参与者均接受了1.5T 3-D T2加权快速自旋回波MRCP检查，分别使用传统序列（三倍加速）和MRCP（八倍加速）。三位盲法阅片者独立评估图像质量，包括背景信号抑制、胆管和胰管可视性、伪影水平以及诊断信心，采用四级有序量表进行评估。使用配对t检验比较采集时间。图像质量参数采用重复测量方差分析进行评估。阅片者间一致性采用Fleiss'κ分析。
在419例连续患者中，评估了30例参与者（平均年龄63±15岁；男性16例，女性14例）。传统MRCP的平均采集时间为10:30±03:04分钟，MRCP为3:57±01:13分钟，P<0.001。MRCP将采集时间缩短了62.4%。传统MRCP的伪影水平评分为3.17±0.77，MRCP为3.56±0.66（P=0.041）。背景信号抑制、胆管可视性、胰管可视性和诊断信心无显著差异（P>0.05）。阅片者间一致性为实质性到几乎完美（κ：0.64-87）。
深度学习加速的3-D MRCP将采集时间缩短了62%，最大限度地减少了伪影，并保留了胆管和胰管的可视性，支持其在常规临床实践中的应用。
