Deep Learning Analysis of Cardiac MRI in Legacy Datasets: Multi-Ethnic Study of Atherosclerosis
The Multi-Ethnic Study of Atherosclerosis (MESA), begun in 2000, was the first large cohort study to incorporate cardiovascular magnetic resonance (CMR) to study the mechanisms of cardiovascular disease in over 5,000 initially asymptomatic participants, and there is now a wealth of follow-up data ov...
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2022-01-01
|
| Series: | Frontiers in Cardiovascular Medicine |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fcvm.2021.807728/full |
| _version_ | 1798014725981732864 |
|---|---|
| author | Avan Suinesiaputra Avan Suinesiaputra Charlène A. Mauger Bharath Ambale-Venkatesh David A. Bluemke Josefine Dam Gade Kathleen Gilbert Markus H. A. Janse Line Sofie Hald Conrad Werkhoven Colin O. Wu Joao A. C. Lima Alistair A. Young |
| author_facet | Avan Suinesiaputra Avan Suinesiaputra Charlène A. Mauger Bharath Ambale-Venkatesh David A. Bluemke Josefine Dam Gade Kathleen Gilbert Markus H. A. Janse Line Sofie Hald Conrad Werkhoven Colin O. Wu Joao A. C. Lima Alistair A. Young |
| author_sort | Avan Suinesiaputra |
| collection | DOAJ |
| description | The Multi-Ethnic Study of Atherosclerosis (MESA), begun in 2000, was the first large cohort study to incorporate cardiovascular magnetic resonance (CMR) to study the mechanisms of cardiovascular disease in over 5,000 initially asymptomatic participants, and there is now a wealth of follow-up data over 20 years. However, the imaging technology used to generate the CMR images is no longer in routine use, and methods trained on modern data fail when applied to such legacy datasets. This study aimed to develop a fully automated CMR analysis pipeline that leverages the ability of machine learning algorithms to enable extraction of additional information from such a large-scale legacy dataset, expanding on the original manual analyses. We combined the original study analyses with new annotations to develop a set of automated methods for customizing 3D left ventricular (LV) shape models to each CMR exam and build a statistical shape atlas. We trained VGGNet convolutional neural networks using a transfer learning sequence between two-chamber, four-chamber, and short-axis MRI views to detect landmarks. A U-Net architecture was used to detect the endocardial and epicardial boundaries in short-axis images. The landmark detection network accurately predicted mitral valve and right ventricular insertion points with average error distance <2.5 mm. The agreement of the network with two observers was excellent (intraclass correlation coefficient >0.9). The segmentation network produced average Dice score of 0.9 for both myocardium and LV cavity. Differences between the manual and automated analyses were small, i.e., <1.0 ± 2.6 mL/m2 for indexed LV volume, 3.0 ± 6.4 g/m2 for indexed LV mass, and 0.6 ± 3.3% for ejection fraction. In an independent atlas validation dataset, the LV atlas built from the fully automated pipeline showed similar statistical relationships to an atlas built from the manual analysis. Hence, the proposed pipeline is not only a promising framework to automatically assess additional measures of ventricular function, but also to study relationships between cardiac morphologies and future cardiac events, in a large-scale population study. |
| first_indexed | 2024-04-11T15:22:49Z |
| format | Article |
| id | doaj.art-cef62ee8450f4a568faa947387f9e59c |
| institution | Directory Open Access Journal |
| issn | 2297-055X |
| language | English |
| last_indexed | 2024-04-11T15:22:49Z |
| publishDate | 2022-01-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Cardiovascular Medicine |
| spelling | doaj.art-cef62ee8450f4a568faa947387f9e59c2022-12-22T04:16:20ZengFrontiers Media S.A.Frontiers in Cardiovascular Medicine2297-055X2022-01-01810.3389/fcvm.2021.807728807728Deep Learning Analysis of Cardiac MRI in Legacy Datasets: Multi-Ethnic Study of AtherosclerosisAvan Suinesiaputra0Avan Suinesiaputra1Charlène A. Mauger2Bharath Ambale-Venkatesh3David A. Bluemke4Josefine Dam Gade5Kathleen Gilbert6Markus H. A. Janse7Line Sofie Hald8Conrad Werkhoven9Colin O. Wu10Joao A. C. Lima11Alistair A. Young12Department of Anatomy and Medical Imaging, University of Auckland, Auckland, New ZealandDepartment of Biomedical Engineering, School of Biomedical Engineering and Imaging Sciences, King's College London, London, United KingdomDepartment of Anatomy and Medical Imaging, University of Auckland, Auckland, New ZealandJohns Hopkins Medical Center, Baltimore, MD, United StatesDepartment of Radiology, University of Wisconsin School of Medicine and Public Health, Madison, WI, United StatesDepartment of Biomedical Engineering and Informatics, School of Medicine and Health, Aalborg University, Aalborg, DenmarkAuckland Bioengineering Institute, University of Auckland, Auckland, New ZealandDepartment of Electrical Engineering, Eindhoven University of Technology, Eindhoven, NetherlandsDepartment of Biomedical Engineering and Informatics, School of Medicine and Health, Aalborg University, Aalborg, DenmarkAuckland Bioengineering Institute, University of Auckland, Auckland, New ZealandDivision of Intramural Research, National Heart, Lung and Blood Institute, National Institutes of Health, Baltimore, MD, United StatesJohns Hopkins Medical Center, Baltimore, MD, United StatesFaculty of Life Sciences & Medicine, School of Biomedical Engineering & Imaging Sciences, King's College London, London, United KingdomThe Multi-Ethnic Study of Atherosclerosis (MESA), begun in 2000, was the first large cohort study to incorporate cardiovascular magnetic resonance (CMR) to study the mechanisms of cardiovascular disease in over 5,000 initially asymptomatic participants, and there is now a wealth of follow-up data over 20 years. However, the imaging technology used to generate the CMR images is no longer in routine use, and methods trained on modern data fail when applied to such legacy datasets. This study aimed to develop a fully automated CMR analysis pipeline that leverages the ability of machine learning algorithms to enable extraction of additional information from such a large-scale legacy dataset, expanding on the original manual analyses. We combined the original study analyses with new annotations to develop a set of automated methods for customizing 3D left ventricular (LV) shape models to each CMR exam and build a statistical shape atlas. We trained VGGNet convolutional neural networks using a transfer learning sequence between two-chamber, four-chamber, and short-axis MRI views to detect landmarks. A U-Net architecture was used to detect the endocardial and epicardial boundaries in short-axis images. The landmark detection network accurately predicted mitral valve and right ventricular insertion points with average error distance <2.5 mm. The agreement of the network with two observers was excellent (intraclass correlation coefficient >0.9). The segmentation network produced average Dice score of 0.9 for both myocardium and LV cavity. Differences between the manual and automated analyses were small, i.e., <1.0 ± 2.6 mL/m2 for indexed LV volume, 3.0 ± 6.4 g/m2 for indexed LV mass, and 0.6 ± 3.3% for ejection fraction. In an independent atlas validation dataset, the LV atlas built from the fully automated pipeline showed similar statistical relationships to an atlas built from the manual analysis. Hence, the proposed pipeline is not only a promising framework to automatically assess additional measures of ventricular function, but also to study relationships between cardiac morphologies and future cardiac events, in a large-scale population study.https://www.frontiersin.org/articles/10.3389/fcvm.2021.807728/fullcardiac anatomymachine learningleft ventricleMRIdeep learning |
| spellingShingle | Avan Suinesiaputra Avan Suinesiaputra Charlène A. Mauger Bharath Ambale-Venkatesh David A. Bluemke Josefine Dam Gade Kathleen Gilbert Markus H. A. Janse Line Sofie Hald Conrad Werkhoven Colin O. Wu Joao A. C. Lima Alistair A. Young Deep Learning Analysis of Cardiac MRI in Legacy Datasets: Multi-Ethnic Study of Atherosclerosis Frontiers in Cardiovascular Medicine cardiac anatomy machine learning left ventricle MRI deep learning |
| title | Deep Learning Analysis of Cardiac MRI in Legacy Datasets: Multi-Ethnic Study of Atherosclerosis |
| title_full | Deep Learning Analysis of Cardiac MRI in Legacy Datasets: Multi-Ethnic Study of Atherosclerosis |
| title_fullStr | Deep Learning Analysis of Cardiac MRI in Legacy Datasets: Multi-Ethnic Study of Atherosclerosis |
| title_full_unstemmed | Deep Learning Analysis of Cardiac MRI in Legacy Datasets: Multi-Ethnic Study of Atherosclerosis |
| title_short | Deep Learning Analysis of Cardiac MRI in Legacy Datasets: Multi-Ethnic Study of Atherosclerosis |
| title_sort | deep learning analysis of cardiac mri in legacy datasets multi ethnic study of atherosclerosis |
| topic | cardiac anatomy machine learning left ventricle MRI deep learning |
| url | https://www.frontiersin.org/articles/10.3389/fcvm.2021.807728/full |
| work_keys_str_mv | AT avansuinesiaputra deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis AT avansuinesiaputra deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis AT charleneamauger deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis AT bharathambalevenkatesh deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis AT davidabluemke deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis AT josefinedamgade deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis AT kathleengilbert deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis AT markushajanse deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis AT linesofiehald deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis AT conradwerkhoven deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis AT colinowu deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis AT joaoaclima deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis AT alistairayoung deeplearninganalysisofcardiacmriinlegacydatasetsmultiethnicstudyofatherosclerosis |