Share：
Share this content in WeChat
X
[Chinese][PDF]818121
• Clinical Article •
The feasibility of semi-automatic segmentation technology for quantification of pancreatic fat: Comparative study with traditional ROI methods
YOU Yaru  ZHANG Qinhe  LIU Ailian  LIANG Chao  WANG Jiazheng  LIN Liangjie  CHEN Lihua  SONG Qingwei 

Cite this article as: You YR, Zhang QH, Liu AL, et al. The feasibility of semi-automatic segmentation technology for quantification of pancreatic fat: Comparative study with traditional ROI methods. Chin J Magn Reson Imaging, 2020, 11(12): 1124-1128. DOI:10.12015/issn.1674-8034.2020.12.009.


[Abstract] Objective: To explore the feasibility of quantifying pancreatic fat fraction (PFF) with semi-automatic segmentation using 3D mDixon Quant sequence.Materials and Methods: Thirty healthy subjects who underwent upper abdomen 3.0 T MR (Ingenia CX, Philips) including 3D mDixon Quant sequences were collected. There were 14 males and 16 females, age range 22—69 years (median age, 48 years), body mass index (BMI) range 17.71—32.59 kg/m2 (median, 24.57 kg/m2). After scanning, the images were imported into ISP (IntelliSpace Portal, Philips) workstation. Traditional regions of interest (ROI) placement and semi-automatic segmentation techniques were used to measure PFF on fat fraction images. Three ROIs were placed on the pancreatic uncinate process, head and neck, body, and tail, avoiding blood vessels, pancreatic duct and visceral adipose tissue. The mean value of ROIs was recorded as the FF in this region, and then the average FF of all regions were calculated as the whole PFF. Pancreatic tissue was manually traced on fat fraction images and was semi-automatically segmented, and then whole PFF was calculated automatically. The operation time was recorded. Two observers (3 and 5 years of imaging diagnosis experience) used two methods to measure PFF. Data were analyzed by SPSS 22.0. Shapiro-Wilk test was used to test the normality of the data. The intra-class correlation coefficient (ICC) was used to test the consistency of the data. The non-parametric Mann-Whitney U test was used to test the differences in PFF and the operation time by the two methods.Results: The data consistency of two observers was good (ICC=0.981, 0.929). The consistency of whole PFF in two groups was good (ICC=0.981). The whole PFF (%) of the two groups were 3.73 (2.97, 5.84) and 4.20 (3.05, 6.21) respectively. The semi-automatic segmentation technique measured the pancreatic fat fraction slightly larger than the traditional ROI placement method, and there was no significant difference (Z=-0.466, P>0.05). The operation time was statistically different (85 s vs 133 s, Z=-6.238, P<0.05).Conclusions: The semi-automatic segmentation technology is feasible for the quantification of pancreatic fat. It can significantly shorten the measurement time under the condition of ensuring the accuracy of the results, and has a good clinical application prospect.
[Keywords] magnetic resonance imaging；fat quantification；pancreas；semi-automatic segmentation；regions of interest

YOU Yaru Department of Radiology, the First Affiliated Hospital of Dalian Medical University, Dalian 116011, China

ZHANG Qinhe Department of Radiology, the First Affiliated Hospital of Dalian Medical University, Dalian 116011, China

LIU Ailian* Department of Radiology, the First Affiliated Hospital of Dalian Medical University, Dalian 116011, China

LIANG Chao Department of Radiology, the First Affiliated Hospital of Dalian Medical University, Dalian 116011, China

WANG Jiazheng MSC Clinical & Technical Solutions, Philips Healthcare, Beijing 100600, China

LIN Liangjie MSC Clinical & Technical Solutions, Philips Healthcare, Beijing 100600, China

CHEN Lihua Department of Radiology, the First Affiliated Hospital of Dalian Medical University, Dalian 116011, China

SONG Qingwei Department of Radiology, the First Affiliated Hospital of Dalian Medical University, Dalian 116011, China

*Correspondence to: Liu AL, E-mail: liuailian@dmu.edu.cn

Conflicts of interest   None.

ACKNOWLEDGMENTS  This work was part of Program for Training Capital Science and Technology Leading Talents No.Z181100006318003
Received  2020-06-29
Accepted  2020-07-29
DOI: 10.12015/issn.1674-8034.2020.12.009
Cite this article as: You YR, Zhang QH, Liu AL, et al. The feasibility of semi-automatic segmentation technology for quantification of pancreatic fat: Comparative study with traditional ROI methods. Chin J Magn Reson Imaging, 2020, 11(12): 1124-1128. DOI:10.12015/issn.1674-8034.2020.12.009.

[1]
袁贝贝，谭玉娥，刘京京,等.非酒精性脂肪性胰腺病的研究现状.临床肝胆病杂志, 2019, 35(5): 1161-1164. DOI: 10.3969/j.issn.1001-5256.2019.05.052.
[2]
Alempijevic T, Dragasevic S, Zec S, et al. Non-alcoholic fatty pancreas disease. Postgraduate Med J, 2017, 93(1098): 226-230. DOI: 10.1136/postgradmedj-2016-134546.
[3]
Yu TY, Wang CY. Impact of non-alcoholic fatty pancreas disease on glucose metabolism. J Diabetes Investig, 2017, 8(6): 735-747. DOI: 10.1111/jdi.12665.
[4]
林海萍，俞世安.脂肪胰诊断及其临床意义的研究进展.中国临床新医学, 2019, 12(6): 679-682. DOI: 10.3969/j.issn.1674-3806.2019.06.28.
[5]
孙金磊，李葆青，王月卿,等.磁共振波谱和mDIXON-Quant评价椎体骨髓脂肪含量的比较.中华骨质疏松和骨矿盐疾病杂志, 2016, 9(4): 396-401. DOI: 10.3969/j.issn.1674-2591.2016.04.010.
[6]
Weng S, Zhou J, Chen X, et al. Prevalence and factors associated with nonalcoholic fatty pancreas disease and its severity in China. Medicine, 2018, 97(26): e11293. DOI: 10.1097/MD.0000000000011293.
[7]
张斯佳.胰腺灶性脂肪浸润的CT表现.中国实用医刊, 2010, 37(6): 89-90. DOI: 10.3760/cma.j.issn.1674-4756.2010.06.057.
[8]
李晓阳，邢国凤，赵林,等.胰腺脂肪浸润的CT诊断及临床相关性分析.医学影像学杂志, 2010, 20(10): 1475-1477.
[9]
Van Geenen EJ, Smits MM, Schreuder TC, et al. Nonalcoholic fatty liver is related to nonalcoholic fatty pancreas disease. Pancreas, 2010, 39(8): 1185-1190. DOI: 10.1097/MPA.0b013e3181f6fce2.
[10]
Patel NS, Peterson MR, Brenner DA, et al. Association between novel MRI-estimated pancreatic fat and liver histology-determined steatosis and fibrosis in non-alcoholic fatty liver disease. Aliment Pharmacol Ther, 2013, 37(6): 630-639. DOI: 10.1111/apt.12237.
[11]
向佳兵，高媛.超声评价非酒精性脂肪肝患者胰腺脂肪浸润与亚临床动脉粥样硬化的相关性研究.河北医药, 2019, 41(23): 3531-3534. DOI: 10.3969/j.issn.1002-7386.2019.23.002.
[12]
张颖，钱庭月，孙红光.胰腺脂肪定量成像技术的研究进展.中国医学计算机成像杂志, 2019, 25(2): 208-211. DOI: 10.19627/j.cnki.cn31-1700/th.2019.02.023.
[13]
韩广，孟祥福，范欣,等. CT平扫定量测量胰腺脂肪含量的临床应用价值.中国中西医结合影像学杂志, 2018, 16(2): 159-162. DOI: 10.3969/j.issn.1672-0512.2018.02.016.
[14]
Gaborit B, Abdesselam I, Kober F, et al. Ectopic fat storage in the pancreas using 1H-MRS: importance of diabetic status and modulation with bariatric surgery-induced weight loss. Int J Obes, 2015, 39(3): 480-487. DOI: 10.1038/ijo.2014.126.
[15]
Hannukainen JC, Borra R, Linderborg K, et al. Liver and pancreatic fat content and metabolism in healthy monozygotic twins with discordant physical activity. J Hepatol, 2010, 54(3): 545-552. DOI: 10.1016/j.jhep.2010.07.029.
[16]
Szczepaniak LS, Nurenberg P, Leonard D, et al. Magnetic resonance spectroscopy to measure hepatic triglyceride content: prevalence of hepatic steatosis in thegeneral population. Am J Physiol Endocrinol Metab, 2005, 288(2): 462-468. DOI: 10.1152/ajpendo.00064.2004.
[17]
林丽丽，黄子星，宋彬,等.磁共振波谱成像在胰腺癌诊断中的应用进展.中国普外基础与临床杂志, 2013, 20(10): 1179-1182. DOI: 10.7507/1007-9424.20130296.
[18]
聂斌.医学图像分割技术及其进展.泰山医学院学报, 2002, 23(4): 422-426.
[19]
Cho SJ, Choi YJ, Chung SR, et al. High-resolution MRI using compressed sensing-sensitivity encoding (CS-SENSE) for patients with suspected neurovascular compression syndrome: comparison with the conventional SENSE parallel acquisition technique. Clin Radiol, 2019, 74(10): 817-817. DOI: 10.1016/j.crad.2019.06.023.
[20]
Kawamoto S, Siegelman SS, Bluemke DA, et al. Focal fatty infiltration in the head of the pancreas: evaluation with multidetector computed tomography with multiplanar reformation imaging. J Computer Assisted Tomography, 2009, 33(1): 90-95. DOI: 10.1097/rct.0b013e31815cff0d.
[21]
张钦和，刘爱连，宋清伟,等. IDEAL-IQ对非酒精性脂肪肝患者胰腺脂肪含量的定量评估.临床放射学杂志, 2019, 38(3): 553-557. DOI: 10.13437/j.cnki.jcr.2019.03.044.
[22]
Matsumoto S, Mori H, Miyake H, et al. Uneven fatty replacement of the pancreas: evaluation with CT. Radiology, 1995, 194(2): 453-458． DOI: 10.1148/radiology.194.2.7824726.

PREV The value of routine MRI combined with diffusion-weighted imaging in the differential diagnosis of atypical liver abscess and intrahepatic cholangiocarcinoma
NEXT MRI features of the submandibular gland in Sjögren syndrome
  


LINK


Tel & Fax: +8610-67113815    E-mail: editor@cjmri.cn