Share：
Share this content in WeChat
X
[Chinese][PDF]151
• Clinical Article •
Preliminary study on deep learning-based enhanced 3D-MRA segmentation in patients with Budd-Chiari syndrome
ZHU Lei  QIU Xianglong  SHA Juncheng  WANG Chao  GOU Yabo  ZHANG Leiming  ZHANG Qingqiao 

Cite this article as: ZHU L, QIU X L, SHA J C, et al. Preliminary study on deep learning-based enhanced 3D-MRA segmentation in patients with Budd-Chiari syndrome[J]. Chin J Magn Reson Imaging, 2025, 16(4): 54-59, 80. DOI:10.12015/issn.1674-8034.2025.04.009.


[Abstract] Objective To evaluate the segmentation performance of a deep learning (DL) model in the analysis of enhanced three dimensional-magnetic resonance angiography (3D-MRA) images of patients with Budd-Chiari syndrome (BCS), and assess the inter-observer agreement among radiologists in the evaluation of the DL model's segmentation outcomes.Materials and Methods A retrospective analysis was conducted on MRA images from 220 BCS patients. Manual segmentation was performed by two radiologists with 8 and 12 years of experience, respectively. The DL model was trained on the features extracted from these manual segmentations to enable automatic segmentation. The performance of the DL model was assessed using the Dice similarity coefficient (DSC), sensitivity, specificity, and accuracy. Consistency comparison between the segmentation results of different radiologists and the DL model was used by the area under the curve (AUC) of receiver operating characteristics (ROC). Inter-observer agreement regarding the DL model's segmentation results was evaluated using the intra-class correlation coefficient (ICC) and Wilcoxon paired test.Results The DL model achieved DSC values of 0.93, 0.84, and 0.65 for the liver, inferior vena cava, and hepatic veins, respectively; sensitivity values were 92%, 81%, and 73%; specificity values were 93%, 93%, and 76%; accuracy values were 95%, 94%, and 86%, and AUC values were 0.95, 0.87, 0.71, respectively. There was no statistically significant difference (P > 0.05) in AUC values between two radiologists and DL model in liver and vascular recognition of BCS patients. The subjective assessments of the DL model's segmentation results by the two radiologists showed no statistically significant differences (P > 0.05). The overall ICC was 0.94 (95% CI: 0.92 to 0.95).Conclusions The DL model exhibited robust segmentation performance in enhanced 3D-MRA images of BCS patients. Furthermore, there was excellent inter-observer agreement among radiologists of the images segmented by the DL model.
[Keywords] deep learning；Budd-Chiari syndrome；magnetic resonance angiography；magnetic resonance imaging；segmentation

ZHU Lei1   QIU Xianglong1   SHA Juncheng1   WANG Chao1   GOU Yabo1   ZHANG Leiming2   ZHANG Qingqiao1*  

1 Department of Interventional Radiology, the Affiliated Hospital of Xuzhou Medical University, Budd-Chiari Syndrome Diagnosis and Treatment Center of Jiangsu Province, Xuzhou 221006, China

2 Intervention Center of the First Affiliated Hospital of Bengbu Medical University, Bengbu 233004, China

Corresponding author: ZHANG Q Q, E-mail: 1427286069@qq.com

Conflicts of interest   None.

Received  2024-11-21
Accepted  2025-04-10
DOI: 10.12015/issn.1674-8034.2025.04.009
Cite this article as: ZHU L, QIU X L, SHA J C, et al. Preliminary study on deep learning-based enhanced 3D-MRA segmentation in patients with Budd-Chiari syndrome[J]. Chin J Magn Reson Imaging, 2025, 16(4): 54-59, 80. DOI:10.12015/issn.1674-8034.2025.04.009.

[1]
SHUKLA A, SHRESHTHA A, MUKUND A, et al. Budd-Chiari syndrome: consensus guidance of the Asian Pacific Association for the study of the liver (APASL)[J]. Hepatol Int, 2021, 15(3): 531-567. DOI: 10.1007/s12072-021-10189-4.
[2]
GARCIA-PAGÁN J C, VALLA D C. Primary budd-chiari syndrome[J]. N Engl J Med, 2023, 388(14): 1307-1316. DOI: 10.1056/nejmra2207738.
[3]
LI Y Y, DE STEFANO V, LI H Y, et al. Epidemiology of budd-chiari syndrome: a systematic review and meta-analysis[J]. Clin Res Hepatol Gastroenterol, 2019, 43(4): 468-474. DOI: 10.1016/j.clinre.2018.10.014.
[4]
WANG X, ZHOU X Y, GAO C, et al. Analysis of 169 cases of Budd-Chiari syndrome in Shanxi province[J]. J Shanxi Med Univ, 2014, 45(6): 481-483. DOI: 10.13753/j.issn.1007-6611.2014.06.012.
[5]
XU P, LYU L L, SAMI M U, et al. Diagnostic accuracy of magnetic resonance angiography for Budd-Chiari syndrome: A meta-analysis[J]. Exp Ther Med, 2018, 16(6): 4873-4878. DOI: 10.3892/etm.2018.6764.
[6]
XU P, LYU L L, LU X, et al. Evaluating the short-term clinical efficacy of magnetic resonance elastography in patients with budd-chiari syndrome[J]. Acad Radiol, 2021, 28(Suppl 1): S179-S183. DOI: 10.1016/j.acra.2021.02.016.
[7]
LIN X H, WU S B, LI L, et al. Automatic mammographic breast density classification in Chinese women: clinical validation of a deep learning model[J]. Acta Radiol, 2023, 64(5): 1823-1830. DOI: 10.1177/02841851231152097.
[8]
NEWMAN E, HORESH L, AVRON H, et al. Stable tensor neural networks for efficient deep learning[J/OL]. Front Big Data, 2024, 7: 1363978 [2024-11-25]. https://pubmed.ncbi.nlm.nih.gov/38873283/. DOI: 10.3389/fdata.2024.1363978.
[9]
CHEN Y D, JIANG H Y, CHEN J, et al. Opportunities and challenges of liver cancer imaging: Achievements and prospects over the past decade in China[J]. Chin J Magn Reson Imag, 2022, 13(10): 71-78. DOI: 10.12015/issn.1674-8034.2022.10.010.
[10]
YAN Q S, WANG B, ZHANG W, et al. Attention-guided deep neural network with multi-scale feature fusion for liver vessel segmentation[J]. IEEE J Biomed Health Inform, 2021, 25(7): 2629-2642. DOI: 10.1109/JBHI.2020.3042069.
[11]
Expert committee on vane cava obstruction of specialized committee of endovascology of chinese medical doctor association. Expert consensus on the classification of subtype in Budd-Chiari syndrome[J]. J Clin Hepatol, 2017, 33(7): 1229-1235. DOI: 10.3969/j.issn.1001-5256.2017.07.005.
[12]
AMLA Z, KHEHRA P S, MATHIALAGAN A, et al. Review of the free research software for computer-assisted interventions[J]. J Imaging Inform Med, 2024, 37(1): 386-401. DOI: 10.1007/s10278-023-00912-y.
[13]
TONG G Y, JIANG H Y, YAO Y D. SDA-UNet: a hepatic vein segmentation network based on the spatial distribution and density awareness of blood vessels[J/OL]. Phys Med Biol, 2023, 68(3) [2024-11-22]. https://pubmed.ncbi.nlm.nih.gov/36623320/. DOI: 10.1088/1361-6560/acb199.
[14]
HAO W, ZHANG J, SU J, et al. HPM-Net: Hierarchical progressive multiscale network for liver vessel segmentation in CT images[J/OL]. Comput Methods Programs Biomed, 2022, 224: 107003 [2024-11-20]. https://pubmed.ncbi.nlm.nih.gov/35868034/. DOI: 10.1016/j.cmpb.2022.107003.
[15]
GOLLA A K, BAUER D F, SCHMIDT R, et al. Convolutional neural network ensemble segmentation with ratio-based sampling for the arteries and veins in abdominal CT scans[J]. IEEE Trans Biomed Eng, 2021, 68(5): 1518-1526. DOI: 10.1109/TBME.2020.3042640.
[16]
YU J X, WU C Y, LI Y Y, et al. Intelligent identification of coal crack in CT images based on deep learning[J/OL]. Comput Intell Neurosci, 2022, 2022: 7092436 [2024-11-21]. https://pubmed.ncbi.nlm.nih.gov/36193183/. DOI: 10.1155/2022/7092436.
[17]
CAYOT B, MILOT L, NEMPONT O, et al. Polycystic liver: automatic segmentation using deep learning on CT is faster and as accurate compared to manual segmentation[J]. Eur Radiol, 2022, 32(7): 4780-4790. DOI: 10.1007/s00330-022-08549-1.
[18]
RIZZETTO F, RUTANNI D, CARBONARO L A, et al. Focal liver lesions in budd-chiari syndrome: spectrum of imaging findings[J/OL]. Diagnostics, 2023, 13(14): 2346 [2024-10-28]. https://pubmed.ncbi.nlm.nih.gov/37510090/. DOI: 10.3390/diagnostics13142346.
[19]
SENGUN B, ISCAN Y, YAZICI Z A, et al. Utilization of artificial intelligence in minimally invasive right adrenalectomy: recognition of anatomical landmarks with deep learning[J]. Acta Chir Belg, 2024, 124(6): 492-498. DOI: 10.1080/00015458.2024.2363599.
[20]
NAI R L, LIN Z Y, E·T Y, et al. Exploring 3D MRU urinary tract segmentation based on deep learning[J]. Radiol Pract, 2022, 37(7): 865-869. DOI: 10.13609/j.cnki.1000-0313.2022.07.012..
[21]
HUANG Q, SUN J F, DING H, et al. Robust liver vessel extraction using 3D U-Net with variant dice loss function[J]. Comput Biol Med, 2018, 101: 153-162. DOI: 10.1016/j.compbiomed.2018.08.018.
[22]
ZBINDEN L, CATUCCI D, SUTER Y, et al. Convolutional neural network for automated segmentation of the liver and its vessels on non-contrast T1 vibe Dixon acquisitions[J/OL]. Sci Rep, 2022, 12(1): 22059 [2024-10-15]. https://pubmed.ncbi.nlm.nih.gov/36543852/. DOI: 10.1038/s41598-022-26328-2.
[23]
LUO T, GAO Y, WU Q, et al. Meta analysis of the diagnostic value of carotid artery high-resolution magnetic resonance vessel wall imaging in the occurrence and recurrence of ischemic stroke[J]. Chin J Magn Reson Imag, 2024, 15(7): 70-75. DOI: 10.12015/issn.1674-8034.2024.07.012.
[24]
LIN A, MANRAL N, MCELHINNEY P, et al. Deep learning-enabled coronary CT angiography for plaque and stenosis quantification and cardiac risk prediction: an international multicentre study[J/OL]. 2022, 4(4): e256-e265 [2024-11-22]. https://pubmed.ncbi.nlm.nih.gov/35337643/. DOI: 10.1016/S2589-7500(22)00022-X.
[25]
LIN A, MANRAL N, MCELHINNEY P, et al. Deep learning-based plaque quantification from coronary computed tomography angiography: external validation and comparison with intravascular ultrasound[J/OL]. Eur Heart J, 2021, 42(Supplement_1): ehab724.0161 [2024-11-22]. https://academic.oup.com/eurheartj/article/42/Supplement_1/ehab724.0161/6392427. DOI: 10.1093/eurheartj/ehab724.0161.
[26]
FAN X X, LIU K, CHE Y, et al. Good clinical outcomes in budd-chiari syndrome with hepatic vein occlusion[J]. Dig Dis Sci, 2016, 61(10): 3054-3060. DOI: 10.1007/s10620-016-4208-0.
[27]
LIN M, ZHANG F, WANG Y, et al. Liver cirrhosis caused by chronic Budd-Chiari syndrome[J/OL]. Medicine (Baltimore), 2017, 96(34): e7425 [2024-11-19]. https://pubmed.ncbi.nlm.nih.gov/28834866/. DOI: 10.1097/md.0000000000007425.
[28]
QI X, GUO X, FAN D. Difference in budd-chiari syndrome between the west and China[J/OL]. Hepatology, 2015, 62(2): 656 [2024-11-16]. https://pubmed.ncbi.nlm.nih.gov/25476543/. DOI: 10.1002/hep.27628.
[29]
MANCUSO A. Budd-chiari syndrome in the west and the east: same syndrome, different diseases[J/OL]. Liver Int, 2019, 39(12): 2417 [2024-10-18]. https://pubmed.ncbi.nlm.nih.gov/31066181/. DOI: 10.1111/liv.14131.
[30]
ZU M H, XU H, ZHANG Q Q, et al. Review of budd-chiari syndrome[J]. J Interv Med, 2020, 3(2): 65-76. DOI: 10.1016/j.jimed.2020.03.002.

PREV Preliminary study of three-dimensions amide proton metastasis imaging in the prediction of pathological grade of hepatocellular carcinoma
NEXT Assessing the efficacy of MRI radiomics for KRAS mutation prediction in colorectal cancer: insights from a systematic review and Meta-analysis
  


LINK


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