본문 바로가기

Automatic segmentation of multiple cardiovascular structures from cardiac computed tomography angiography images using deep learning

문서 유형
연구출판물 · 학술지 논문
등재
SCIEScopus
저널 순위
JCR Q2Scopus Q1
저널
PLOS ONE15(5) · 2020
피인용
Web of Science 32Scopus 42
FWCI (Scopus)
0
JCR IF
JCR 2019 2.74
ISSN
1932-6203

저자

  1. Baskaran, Lohendran제1저자교신저자Dalio Institute of Cardiovascular Imaging, Weill Cornell Medicine, New York, New York, United States of America,
  2. Al’Aref, Subhi J.Dalio Institute of Cardiovascular Imaging, Weill Cornell Medicine, New York, New York, United States of America,
  3. Maliakal, GabrielCleerly, Inc, New York, New York, United States of America,
  4. Lee, Benjamin C.Dalio Institute of Cardiovascular Imaging, Weill Cornell Medicine, New York, New York, United States of America,
  5. Xu, ZhuoranDalio Institute of Cardiovascular Imaging, Weill Cornell Medicine, New York, New York, United States of America,
  6. Choi, Jeong W.Dalio Institute of Cardiovascular Imaging, Weill Cornell Medicine, New York, New York, United States of America,
  7. Lee, Sang-EunDivision of Cardiology, Severance Cardiovascular Hospital, Integrative Cardiovascular Imaging Center, Yonsei University College of Medicine, Seoul, South Korea,
  8. Sung, Ji MinDivision of Cardiology, Severance Cardiovascular Hospital, Integrative Cardiovascular Imaging Center, Yonsei University College of Medicine, Seoul, South Korea,
  9. Lin, Fay Y.Dalio Institute of Cardiovascular Imaging, Weill Cornell Medicine, New York, New York, United States of America,
  10. Dunham, SimonDalio Institute of Cardiovascular Imaging, Weill Cornell Medicine, New York, New York, United States of America,
  11. Mosadegh, BobakDalio Institute of Cardiovascular Imaging, Weill Cornell Medicine, New York, New York, United States of America,
  12. Kim, Yong-JinDepartment of Internal Medicine, Seoul National University College of Medicine, Cardiovascular Center, Seoul National University Hospital, Seoul, South Korea,
  13. Gottlieb, IlanDepartment of Radiology, Casa de Saude São Jose, Rio de Janeiro, Brazil,
  14. Lee, Byoung KwonDivision of Cardiology, Gangnam Severance Hospital, Yonsei University College of Medicine, Seoul, Korea,
  15. Chun, Eun JuDepartment of Radiology, Seoul National University Bundang Hospital, Sungnam, South Korea,
  16. Cademartiri, FilippoCardiovascular Imaging Center, SDN IRCCS, Naples, Italy,
  17. Maffei, EricaDepartment of Radiology, Area Vasta 1/ ASUR Marche, Urbino, Italy,
  18. Marques, HugoUNICA, Unit of Cardiovascular Imaging, Hospital da Luz, Lisboa, Portugal,
  19. Shin, SanghoonDivision of Cardiology, Department of Internal Medicine, Ewha Womans University Seoul Hospital, Seoul, Korea,
  20. 최정현Pusan University Hospital, Busan, South Korea,
  21. Chinnaiyan, KavithaDepartment of Cardiology, William Beaumont Hospital, Royal Oak, Michigan, United States of America,
  22. Hadamitzky, MartinDepartment of Radiology and Nuclear Medicine, German Heart Center Munich, Munich, Germany,
  23. Conte, EdoardoCentro Cardiologico Monzino, IRCCS, Milan, Italy,
  24. Andreini, DanieleCentro Cardiologico Monzino, IRCCS, Milan, Italy,
  25. Pontone, GianlucaCentro Cardiologico Monzino, IRCCS, Milan, Italy,
  26. Budoff, Matthew J.Department of Medicine, Los Angeles Biomedical Research Institute, Torrance, California, United States of America,
  27. Leipsic, Jonathon A.Department of Medicine and Radiology, University of British Columbia, Vancouver, British Columbia, Canada,
  28. Raff, Gilbert L.Department of Cardiology, William Beaumont Hospital, Royal Oak, Michigan, United States of America,
  29. Virmani, RenuDepartment of Pathology, CVPath Institute, Gaithersburg, Maryland, United States of America,
  30. Samady, HabibDivision of Cardiology, Emory University School of Medicine, Atlanta, Georgia, United States of America,
저자 37명 모두 보기

초록

Background Segmentation of cardiovascular images is resource-intensive. We design an automated deep learning method for the segmentation of multiple structures from Coronary Computed Tomography Angiography (CCTA) images. Methods Images from a multicenter registry of patients that underwent clinically-indicated CCTA were used. The proximal ascending and descending aorta (PAA, DA), superior and inferior vena cavae (SVC, IVC), pulmonary artery (PA), coronary sinus (CS), right ventricular wall (RVW) and left atrial wall (LAW) were annotated as ground truth. The U-net-derived deep learning model was trained, validated and tested in a 70:20:10 split. Results The dataset comprised 206 patients, with 5.130 billion pixels. Mean age was 59.9 +/- 9.4 yrs., and was 42.7% female. An overall median Dice score of 0.820 (0.782, 0.843) was achieved. Median Dice scores for PAA, DA, SVC, IVC, PA, CS, RVW and LAW were 0.969 (0.979, 0.988), 0.953 (0.955, 0.983), 0.937 (0.934, 0.965), 0.903 (0.897, 0.948), 0.775 (0.724, 0.925), 0.720 (0.642, 0.809), 0.685 (0.631, 0.761) and 0.625 (0.596, 0.749) respectively. Apart from the CS, there were no significant differences in performance between sexes or age groups. Conclusions An automated deep learning model demonstrated segmentation of multiple cardiovascular structures from CCTA images with reasonable overall accuracy when evaluated on a pixel level.

키워드

HEART

식별자

Scopus EID
2-s2.0-85084406001
WoS UID
000537283500043