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

被引：23

作者：

Baskaran, Lohendran ^{[1
,2
,3
,4
]}

Al'Aref, Subhi J. ^{[1
,2
,3
]}

Maliakal, Gabriel ^{[5
]}

Lee, Benjamin C. ^{[1
]}

Xu, Zhuoran ^{[1
]}

Choi, Jeong W. ^{[1
]}

Lee, Sang-Eun ^{[6
,7
]}

Sung, Ji Min ^{[6
]}

Lin, Fay Y. ^{[1
,2
,3
]}

Dunham, Simon ^{[1
]}

Mosadegh, Bobak ^{[1
]}

Kim, Yong-Jin ^{[8
]}

Gottlieb, Ilan ^{[9
]}

Lee, Byoung Kwon ^{[10
]}

Chun, Eun Ju ^{[11
]}

Cademartiri, Filippo ^{[12
]}

Maffei, Erica ^{[13
]}

Marques, Hugo ^{[14
]}

Shin, Sanghoon ^{[7
]}

Choi, Jung Hyun ^{[15
]}

Chinnaiyan, Kavitha ^{[16
]}

Hadamitzky, Martin ^{[17
]}

Conte, Edoardo ^{[18
]}

Andreini, Daniele ^{[18
]}

Pontone, Gianluca ^{[18
]}

Budoff, Matthew J. ^{[19
]}

Leipsic, Jonathon A. ^{[20
]}

Raff, Gilbert L. ^{[16
]}

Virmani, Renu ^{[16
,21
]}

Samady, Habib ^{[22
]}

Stone, Peter H. ^{[23
]}

Berman, Daniel S. ^{[24
]}

Narula, Jagat ^{[25
,26
]}

Bax, Jeroen J. ^{[27
]}

Chang, Hyuk-Jae ^{[6
]}

Min, James K. ^{[1
,2
,3
]}

Shaw, Leslee J. ^{[1
,2
,3
]}

机构：

[1] Weill Cornell Med, Dalio Inst Cardiovasc Imaging, New York, NY 10065 USA

[2] New York Presbyterian Hosp, Dept Radiol, New York, NY 10038 USA

[3] Weill Cornell Med, New York, NY 10065 USA

[4] Natl Heart Ctr, Dept Cardiovasc Med, Singapore, Singapore

[5] Cleerly Inc, New York, NY USA

[6] Yonsei Univ, Integrat Cardiovasc Imaging Ctr, Severance Cardiovasc Hosp, Div Cardiol,Coll Med, Seoul, South Korea

[7] Ewha Womans Univ, Dept Internal Med, Div Cardiol, Seoul Hosp, Seoul, South Korea

[8] Seoul Natl Univ, Seoul Natl Univ Hosp, Cardiovasc Ctr, Dept Internal Med,Coll Med, Seoul, South Korea

[9] Casa Saude Sao Jose, Dept Radiol, Rio De Janeiro, Brazil

[10] Yonsei Univ, Gangnam Severance Hosp, Div Cardiol, Coll Med, Seoul, South Korea

[11] Seoul Natl Univ, Dept Radiol, Bundang Hosp, Sungnam, South Korea

[12] SDN IRCCS, Cardiovasc Imaging Ctr, Naples, Italy

[13] ASUR Marche, Dept Radiol, Area Vasta 1, Urbino, Italy

[14] Hosp Luz, Unit Cardiovasc Imaging, UNICA, Lisbon, Portugal

[15] Pusan Univ Hosp, Busan, South Korea

[16] William Beaumont Hosp, Dept Cardiol, Royal Oak, MI 48072 USA

[17] German Heart Ctr Munich, Dept Radiol & Nucl Med, Munich, Germany

[18] IRCCS, Ctr Cardiol Monzino, Milan, Italy

[19] Los Angeles Biomed Res Inst, Dept Med, Torrance, CA USA

[20] Univ British Columbia, Dept Med & Radiol, Vancouver, BC, Canada

[21] CVPath Inst, Dept Pathol, Gaithersburg, MD USA

[22] Emory Univ, Sch Med, Div Cardiol, Atlanta, GA 30322 USA

[23] Harvard Med Sch, Brigham & Womens Hosp, Cardiovasc Div, Boston, MA 02115 USA

[24] Cedars Sinai Med Ctr, Dept Imaging & Med, Los Angeles, CA 90048 USA

[25] Icahn Sch Med Mt Sinai, Mt Sinai Heart Zena & Michael Wiener Cardiovasc I, New York, NY 10029 USA

[26] Icahn Sch Med Mt Sinai, Henry R Kravis Ctr Cardiovasc Hlth, New York, NY 10029 USA

[27] Leiden Univ, Dept Cardiol, Med Ctr, Leiden, Netherlands

来源：

PLOS ONE | 2020年 / 15卷 / 05期

基金：

美国国家卫生研究院;

关键词：

HEART;

D O I：

10.1371/journal.pone.0232573

中图分类号：

O [数理科学和化学]; P [天文学、地球科学]; Q [生物科学]; N [自然科学总论];

学科分类号：

07 ; 0710 ; 09 ;

摘要：

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.

引用

页数：13

共 50 条

[1] AUTOMATIC SEGMENTATION OF CARDIOVASCULAR STRUCTURES IMAGED ON CARDIAC COMPUTED TOMOGRAPHY ANGIOGRAPHY USING DEEP LEARNING
Baskaran, Lohendran
Singh, Gurpreet
Xu, Zhuoran
Lee, Benjamin
Choi, Jeong
Gianni, Umberto
van Rosendael, Alexander
van den Hoogen, Inge J.
Dunham, Simon
Mosadegh, Bobak
Lin, Fay
Chang, Hyuk-Jae
Min, James
Shaw, Leslee J.
Al'Aref, Subhi
[J]. JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY, 2020, 75 (11) : 3497 - 3497
[2] DEEP LEARNING BASED AUTOMATIC SEGMENTATION OF CARDIAC COMPUTED TOMOGRAPHY
Singh, Gurpreet
Alaref, Subhi
Maliakal, Gabriel
Pandey, Mohit
van Rosendael, Alexander
Lee, Benjamin
Wang, Jing
Xu, Zhouran
Min, James
[J]. JOURNAL OF THE AMERICAN COLLEGE OF CARDIOLOGY, 2019, 73 (09) : 1643 - 1643
[3] Automatic Segmentation of the Left Atrium from Computed Tomography Angiography Images
Kazi, Amaan
Betko, Sage
Salvi, Anish
Menon, Prahlad G.
[J]. ANNALS OF BIOMEDICAL ENGINEERING, 2023, 51 (08) : 1713 - 1722
[4] Automatic Segmentation of the Left Atrium from Computed Tomography Angiography Images
Amaan Kazi
Sage Betko
Anish Salvi
Prahlad G. Menon
[J]. Annals of Biomedical Engineering, 2023, 51 : 1713 - 1722
[5] Liver segmentation from computed tomography images using cascade deep learning
Araujo, Jose Denes Lima
da Cruz, Luana Batista
Diniz, Joao Otavio Bandeira
Ferreira, Jonnison Lima
Silva, Aristofanes Correa
de Paiva, Anselmo Cardoso
Gattass, Marcelo
[J]. COMPUTERS IN BIOLOGY AND MEDICINE, 2022, 140
[6] Automatic Segmentation of Type A Aortic Dissection on Computed Tomography Images Using Deep Learning Approach
Guo, Xiaoya
Liu, Tianshu
Yang, Yi
Dai, Jianxin
Wang, Liang
Tang, Dalin
Sun, Haoliang
[J]. DIAGNOSTICS, 2024, 14 (13)
[7] AB-ResUNet plus : Improving Multiple Cardiovascular Structure Segmentation from Computed Tomography Angiography Images
Habijan, Marija
Galic, Irena
Romic, Kresimir
Leventic, Hrvoje
[J]. APPLIED SCIENCES-BASEL, 2022, 12 (06):
[8] Deep-Learning-Based Automatic Segmentation of Parotid Gland on Computed Tomography Images
Onder, Merve
Evli, Cengiz
Tuerk, Ezgi
Kazan, Orhan
Bayrakdar, Ibrahim Sevki
Celik, Ozer
Costa, Andre Luiz Ferreira
Gomes, Joao Pedro Perez
Ogawa, Celso Massahiro
Jagtap, Rohan
Orhan, Kaan
[J]. DIAGNOSTICS, 2023, 13 (04)
[9] Automatic segmentation of the thoracic aorta in cardiac computed tomography images
Vera, Miguel
Huerfano, Yoleidy
Contreras, Julio
Vera, Maria
Del Mar, Atilio
Chacon, Jose
Wilches-Duran, Sandra
Graterol-Rivas, Modesto
Riano-Wilches, Daniela
Rojas, Joselyn
Bermudez, Valmore
[J]. REVISTA LATINOAMERICANA DE HIPERTENSION, 2016, 11 (04): : 110 - 116
[10] Deep Learning for Detection of Intracranial Aneurysms from Computed Tomography Angiography Images
Liu, Xiujuan
Mao, Jun
Sun, Ning
Yu, Xiangrong
Chai, Lei
Tian, Ye
Wang, Jianming
Liang, Jianchao
Tao, Haiquan
Yuan, Lihua
Lu, Jiaming
Wang, Yang
Zhang, Bing
Wu, Kaihua
Wang, Yiding
Chen, Mengjiao
Wang, Zhishun
Lu, Ligong
[J]. JOURNAL OF DIGITAL IMAGING, 2023, 36 (01) : 114 - 123

← 1 2 3 4 5 →