广东工业大学学报 ›› 2023, Vol. 40 ›› Issue (05): 34-40.doi: 10.12052/gdutxb.220171

• 计算机科学与技术 • 上一篇    下一篇

基于Swin-Unet的主动脉再缩窄预测研究

甘孟坤1, 曾安1, 张小波2   

  1. 1. 广东工业大学 计算机学院, 广东 广州 510006;
    2. 广东工业大学 自动化学院, 广东 广州 510006
  • 收稿日期:2022-11-18 出版日期:2023-09-25 发布日期:2023-09-26
  • 通信作者: 曾安(1978-),女,教授,博士,主要研究方向为图像处理、数据挖掘,E-mail:zengan@gdut.edu.cn
  • 作者简介:甘孟坤(1995-),男,硕士研究生,主要研究方向为医学图像处理
  • 基金资助:
    广东省重点领域研发计划项目(2021B0101220006)；广东省科技计划项目(2019A050510041)；广东省自然科学基金资助项目(2021A1515012300)；云南省重大科技专项项目(202102AA100012)

Aortic Re-coarctation Prediction Research Based on Swin-Unet

Gan Meng-kun1, Zeng An1, Zhang Xiao-bo2   

  1. 1. School of Computer Science and Technology, Guangdong University of Technology, Guangzhou 510006, China;
    2. School of Automation, Guangdong University of Technology, Guangzhou 510006, China
  • Received:2022-11-18 Online:2023-09-25 Published:2023-09-26

HTML

PDF (PC)

542

摘要： 主动脉缩窄(Coarctation of Aorta，CoA) 是主动脉弓部的先天性畸形，自然预后差，需要早期干预甚至急诊手术治疗，同时术后主动脉再缩窄仍是可能面临的问题。目前主动脉再缩窄的预测主要由医生基于病人的临床特征并结合超声心动图(Ultra Sound Cardiogram) 数据展开风险因素分析，结果往往依赖于超声心动图片拍摄质量以及医生诊断经验，诊断误诊较多。本文以患者心脏部位电子计算机断层扫描(Computed Tomography，CT) 的图像结合患者的临床数据为研究对象，提出了一种基于Swin-Unet网络的多模态数据检测框架，框架结合了Swin-Unet网络模型与机器学习模型，展开多模态特征融合分析，实现主动脉再缩窄的早期检测。临床数据集实验结果显示，与传统的仅使用临床数据的预测方法相比，本文方法有效提升了主动脉再缩窄预测的效果，并验证了与再缩窄有关的风险因素，可为临床医学提供参考。

关键词: 主动脉缩窄, 多模态特征融合, 图像分割

Abstract: Coarctation of aorta (CoA) is a congenital malformation of the aortic arch with a poor natural prognosis, which requires early intervention and even emergency surgery. Meanwhile, postoperative aortic re-coarctation is still a possible problem. At present, the prediction of aortic re-coarctation is mainly carried out based on the risk factor analysis of doctors on the clinical characteristics of patients combining with echocardiography (Ultra Sound Cardiogram) data, which is easy to be misdiagnosed. In this paper, a multimodal data detection framework based on Swin-Unet network is proposed based on the images of the patient's heart from computed tomography (CT) combining with the patient's clinical data. The framework carries out multimodal feature fusion analysis by combining the Swin-Unet network and the machine learning models, aiming to perform early detection of aortic re-coarctation. The experimental results on the clinical dataset show that our proposed methodeffectively improves the prediction effect of aortic re-coarctation when compared with the traditional prediction methods using clinical data. Particularly, we verifie the risk factors related to re-coarctation, the results of which provides a reference for clinical medicine.

Key words: coarctation of aorta, multimodal feature fusion, image segmentation

中图分类号: 

  • TP391
[1] BACKER C L, MAVROUDIS C. Congenital heart surgery nomenclature and database project: patent ductus arteriosus, coarctation of the aorta, interrupted aortic arch [J]. The Annals of Thoracic Surgery, 2000, 69(3): 298-307.
[2] KOUCHOUKOS N T, BLACKSTONE E H, HANLEY F L, et al. Kirklin/Barratt-boyes cardiac surgery e-book[M]. Philadelphia: Elsevier Health Sciences, 2012: 353-462.
[3] NANCE J W, RINGEL R E, FISHMAN E K. Coarctation of the aorta in adolescents and adults: a review of clinical features and CT imaging [J]. Journal of Cardiovascular Computed Tomography, 2016, 10(1): 1-12.
[4] 丁文祥, 苏肇伉. 现代小儿心脏外科学[M]. 济南: 山东科学技术出版社, 2013: 399-405.
[5] ERBEL R, ABOYANS V, BOILEAU C, et al. 2014 ESC guidelines on the diagnosis and treatment of aortic diseases [J]. Kardiologia Polska (Polish Heart Journal), 2014, 72(12): 1169-1252.
[6] LOUIS J D, HARVEY B A, MENK J S, et al. Mortality and operative management for patients undergoing repair of coarctation of the aorta: a retrospective review of the pediatric cardiac care consortium [J]. World Journal for Pediatric and Congenital Heart Surgery, 2015, 6(3): 431-437.
[7] SAXENA A. Recurrent coarctation: interventional techniques and results [J]. World Journal for Pediatric and Congenital Heart Surgery, 2015, 6(2): 257-265.
[8] GRØNDAHL C, PEDERSEN T A L, HJORTDAL V E. The medium-term effects of treatment for mild aortic recoarctation [J]. World Journal for Pediatric and Congenital Heart Surgery, 2017, 8(1): 55-61.
[9] LEHNERT A, VILLEMAIN O, GAUDIN R, et al. Risk factors of mortality and recoarctation after coarctation repair in infancy [J]. Interactive CardioVascular and Thoracic Surgery, 2019, 29(3): 469-475.
[10] BURCH P T, COWLEY C G, HOLUBKOV R, et al. Coarctation repair in neonates and young infants: is small size or low weight still a risk factor? [J]. The Journal of Thoracic and Cardiovascular Surgery, 2009, 138(3): 547-552.
[11] MCELHINNEY D B, YANG S G, HOGARTY A N, et al. Recurrent arch obstruction after repair of isolated coarctation of the aorta in neonates and young infants: is low weight a risk factor? [J]. The Journal of Thoracic and Cardiovascular Surgery, 2001, 122(5): 883-890.
[12] TRUONG D T, TANI L Y, MINICH L L A, et al. Factors associated with recoarctation after surgical repair of coarctation of the aorta by way of thoracotomy in young infants [J]. Pediatric Cardiology, 2014, 35(1): 164-170.
[13] FIORE A C, FISCHER L K, SCHWARTZ T, et al. Comparison of angioplasty and surgery for neonatal aortic coarctation [J]. The Annals of Thoracic Surgery, 2005, 80(5): 1659-1665.
[14] GORBATYKH A V, SOYNOV I A, NICHAI N R, et al. Risk factors for aortic coarctation development in young children [J]. Pediatria. Journal Named after G N Speransky, 2017, 96(3): 118-124.
[15] ADAMSON G, KARAMLOU T, MOORE P, et al. Coarctation index predicts recurrent aortic arch obstruction following surgical repair of coarctation of the aorta in infants [J]. Pediatric Cardiology, 2017, 38(6): 1241-1246.
[16] CHIU H H, WANG J K, CHEN Y S, et al. Long-term outcomes of native coarctation of the aorta after balloon angioplasty or surgical aortoplasty in newborns and young infants less than 3 months of age [J]. Acta Cardiologica Sinica, 2013, 29(2): 168-174.
[17] QUAEGEBEUR J M, JONAS R A, WEINBERG A D, et al. Outcomes in seriously ill neonates with coarctation of the aorta: a multiinstitutional study [J]. The Journal of Thoracic and Cardiovascular Surgery, 1994, 108(5): 841-854.
[18] SEN S, GARG S, RAO S G, et al. Native aortic coarctation in neonates and infants: immediate and midterm outcomes with balloon angioplasty and surgery [J]. Annals of Pediatric Cardiology, 2018, 11(3): 261-266.
[19] TULZER A, MAIR R, KREUZER M, et al. Outcome of aortic arch reconstruction in infants with coarctation: Importance of operative approach [J]. The Journal of Thoracic and Cardiovascular Surgery, 2016, 152(6): 1506-1513.
[20] LIANG C D, SU W J, CHUNG H T, et al. Balloon angioplasty for native coarctation of the aorta in neonates and infants with congestive heart failure [J]. Pediatrics & Neonatology, 2009, 50(4): 152-157.
[21] CAO L, SHI R, GE Y, et al. Fully automatic segmentation of type B aortic dissection from CTA images enabled by deep learning [J]. European Journal of Radiology, 2019, 121: 108713.
[22] 何书坤, 靳巧锋, 刘天舒, 等. 多模态影像在二叶式主动脉瓣形态学及功能评价中的应用[J]. 华中科技大学学报 (医学版), 2022, 51(1): 129-133.
HE S K, JIN Q F, LIU T S, et al. Application of multimodality imaging in bicuspid aortic valve morphology and functional assessment [J]. Journal of Huazhong University of Science and Technology (Medical Edition), 2022, 51(1): 129-133.
[23] 陈思楷. 应用超声—CT多模态影像融合技术建立3D打印左心模型的方法学研究[D]. 武汉: 武汉大学, 2019.
[24] HAGER A, SCHREIBER C, NÜTZL S, et al. Mortality and restenosis rate of surgical coarctation repair in infancy: a study of 191 patients [J]. Cardiology, 2009, 112(1): 36-41.
[25] 刘芳, 黄国英, 梁雪村, 等. 主动脉缩窄 96 例临床分析[J]. 中华医学杂志, 2006, 86(26): 1854-1856.
[26] 郑景浩, 徐志伟, 苏肇伉, 等. 婴幼儿主动脉缩窄的外科治疗[J]. 中华胸心血管外科杂志, 2007, 23(2): 90-92.
[1] 陈志惠, 汪仁煌, 汪志敏. 一种改进的CV水平集图像分割方法[J]. 广东工业大学学报, 2015, 32(2): 104-108.
[2] 王振友, 郑少杰,何远才,林洁芝,苏娜,龙洁丽. 一种PET图像感兴趣区域体积特征的计算方法[J]. 广东工业大学学报, 2013, 30(3): 65-69.
[3] 刘丹； 汪仁煌； 高延增； 刘洪江； . 基于高斯函数的LED彩色图像分割方法[J]. 广东工业大学学报, 2007, 24(4): 73-75.
[4] 张海舰； 成思源； 骆少明； 丁炜； . 基于动态规划法的B样条主动轮廓模型[J]. 广东工业大学学报, 2005, 22(4): 26-30.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2017 《广东工业大学学报》编辑部
地址：广州市东风东路729号广东工业大学 邮编：510090
电话：020-37626653,020-37626139,020-37626396
邮箱：xbzrb@gdut.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发