Heo Subin, Yun Jihye, Kim Dong Wook, Park Seo Young, Choi Sang Hyun, Kim Kyuwon, Jung Kee Wook, Myung Seung-Jae, Park Seong Ho
Department of Radiology and Research Institute of Radiology, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Republic of Korea.
Department of Statistics and Data Science, Korea National Open University, Seoul, Republic of Korea.
Eur Radiol. 2025 Jun 17. doi: 10.1007/s00330-025-11737-4.
Small bowel motility can be quantified using cine MRI, but the influence of patient and imaging factors on motility scores remains unclear. This study evaluated whether patient and imaging factors affect motility scores derived from deep learning-based segmentation of cine MRI.
Fifty-four patients (mean age 53.6 ± 16.4 years; 34 women) with chronic constipation or suspected colonic pseudo-obstruction who underwent cine MRI covering the entire small bowel between 2022 and 2023 were included. A deep learning algorithm was developed to segment small bowel regions, and motility was quantified with an optical flow-based algorithm, producing a motility score for each slice. Associations of motility scores with patient factors (age, sex, body mass index, symptoms, and bowel distension) and MRI slice-related factors (anatomical location, bowel area, and anteroposterior position) were analyzed using linear mixed models.
Deep learning-based small bowel segmentation achieved a mean volumetric Dice similarity coefficient of 75.4 ± 18.9%, with a manual correction time of 26.5 ± 13.5 s. Median motility scores per patient ranged from 26.4 to 64.4, with an interquartile range of 3.1-26.6. Multivariable analysis revealed that MRI slice-related factors, including anatomical location with mixed ileum and jejunum (β = -4.9; p = 0.01, compared with ileum dominant), bowel area (first order β = -0.2, p < 0.001; second order β = 5.7 × 10, p < 0.001), and anteroposterior position (first order β = -51.5, p < 0.001; second order β = 28.8, p = 0.004) were significantly associated with motility scores. Patient factors showed no association with motility scores.
Small bowel motility scores were significantly associated with MRI slice-related factors. Determining global motility without adjusting for these factors may be limited.
Question Global small bowel motility can be quantified from cine MRI; however, the confounding factors affecting motility scores remain unclear. Findings Motility scores were significantly influenced by MRI slice-related factors, including anatomical location, bowel area, and anteroposterior position. Clinical relevance Adjusting for slice-related factors is essential for accurate interpretation of small bowel motility scores on cine MRI.
可使用电影磁共振成像（cine MRI）对小肠蠕动进行量化，但患者和成像因素对蠕动评分的影响尚不清楚。本研究评估了患者和成像因素是否会影响基于深度学习的cine MRI分割得出的蠕动评分。
纳入了54例在2022年至2023年间接受覆盖整个小肠的cine MRI检查的慢性便秘或疑似结肠假性梗阻患者（平均年龄53.6±16.4岁；34名女性）。开发了一种深度学习算法来分割小肠区域，并使用基于光流的算法对蠕动进行量化，为每个切片生成一个蠕动评分。使用线性混合模型分析蠕动评分与患者因素（年龄、性别、体重指数、症状和肠扩张）以及MRI切片相关因素（解剖位置、肠面积和前后位置）之间的关联。
基于深度学习的小肠分割的平均体积骰子相似系数为75.4±18.9%，人工校正时间为26.5±13.5秒。每位患者的蠕动评分中位数在26.4至64.4之间，四分位间距为3.1 - 26.6。多变量分析显示，MRI切片相关因素，包括回肠和空肠混合的解剖位置（β = -4.9；p = 0.01，与以回肠为主相比）、肠面积（一阶β = -0.2，p < 0.001；二阶β = 5.7×10，p < 0.001）和前后位置（一阶β = -51.5，p < 0.001；二阶β = 28.8，p = 0.004）与蠕动评分显著相关。患者因素与蠕动评分无关联。
小肠蠕动评分与MRI切片相关因素显著相关。在不调整这些因素的情况下确定整体蠕动可能存在局限性。
问题 可从cine MRI量化全球小肠蠕动；然而，影响蠕动评分的混杂因素尚不清楚。发现 蠕动评分受MRI切片相关因素显著影响，包括解剖位置、肠面积和前后位置。临床意义 调整切片相关因素对于准确解读cine MRI上的小肠蠕动评分至关重要。
