Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping
Abstract Background Cardiovascular magnetic resonance (CMR) cine displacement encoding with stimulated echoes (DENSE) measures heart motion by encoding myocardial displacement into the signal phase, facilitating high accuracy and reproducibility of global and segmental myocardial strain and providin...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2021-03-01
|
| Series: | Journal of Cardiovascular Magnetic Resonance |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12968-021-00712-9 |
| _version_ | 1827279288587517952 |
|---|---|
| author | Sona Ghadimi Daniel A. Auger Xue Feng Changyu Sun Craig H. Meyer Kenneth C. Bilchick Jie Jane Cao Andrew D. Scott John N. Oshinski Daniel B. Ennis Frederick H. Epstein |
| author_facet | Sona Ghadimi Daniel A. Auger Xue Feng Changyu Sun Craig H. Meyer Kenneth C. Bilchick Jie Jane Cao Andrew D. Scott John N. Oshinski Daniel B. Ennis Frederick H. Epstein |
| author_sort | Sona Ghadimi |
| collection | DOAJ |
| description | Abstract Background Cardiovascular magnetic resonance (CMR) cine displacement encoding with stimulated echoes (DENSE) measures heart motion by encoding myocardial displacement into the signal phase, facilitating high accuracy and reproducibility of global and segmental myocardial strain and providing benefits in clinical performance. While conventional methods for strain analysis of DENSE images are faster than those for myocardial tagging, they still require manual user assistance. The present study developed and evaluated deep learning methods for fully-automatic DENSE strain analysis. Methods Convolutional neural networks (CNNs) were developed and trained to (a) identify the left-ventricular (LV) epicardial and endocardial borders, (b) identify the anterior right-ventricular (RV)-LV insertion point, and (c) perform phase unwrapping. Subsequent conventional automatic steps were employed to compute strain. The networks were trained using 12,415 short-axis DENSE images from 45 healthy subjects and 19 heart disease patients and were tested using 10,510 images from 25 healthy subjects and 19 patients. Each individual CNN was evaluated, and the end-to-end fully-automatic deep learning pipeline was compared to conventional user-assisted DENSE analysis using linear correlation and Bland Altman analysis of circumferential strain. Results LV myocardial segmentation U-Nets achieved a DICE similarity coefficient of 0.87 ± 0.04, a Hausdorff distance of 2.7 ± 1.0 pixels, and a mean surface distance of 0.41 ± 0.29 pixels in comparison with manual LV myocardial segmentation by an expert. The anterior RV-LV insertion point was detected within 1.38 ± 0.9 pixels compared to manually annotated data. The phase-unwrapping U-Net had similar or lower mean squared error vs. ground-truth data compared to the conventional path-following method for images with typical signal-to-noise ratio (SNR) or low SNR (p < 0.05), respectively. Bland–Altman analyses showed biases of 0.00 ± 0.03 and limits of agreement of − 0.04 to 0.05 or better for deep learning-based fully-automatic global and segmental end-systolic circumferential strain vs. conventional user-assisted methods. Conclusions Deep learning enables fully-automatic global and segmental circumferential strain analysis of DENSE CMR providing excellent agreement with conventional user-assisted methods. Deep learning-based automatic strain analysis may facilitate greater clinical use of DENSE for the quantification of global and segmental strain in patients with cardiac disease. |
| first_indexed | 2024-04-24T08:16:31Z |
| format | Article |
| id | doaj.art-d7d382a285174e479533756f66c12e85 |
| institution | Directory Open Access Journal |
| issn | 1532-429X |
| language | English |
| last_indexed | 2024-04-24T08:16:31Z |
| publishDate | 2021-03-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Journal of Cardiovascular Magnetic Resonance |
| spelling | doaj.art-d7d382a285174e479533756f66c12e852024-04-17T03:27:50ZengElsevierJournal of Cardiovascular Magnetic Resonance1532-429X2021-03-0123111310.1186/s12968-021-00712-9Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrappingSona Ghadimi0Daniel A. Auger1Xue Feng2Changyu Sun3Craig H. Meyer4Kenneth C. Bilchick5Jie Jane Cao6Andrew D. Scott7John N. Oshinski8Daniel B. Ennis9Frederick H. Epstein10Department of Biomedical Engineering, University of VirginiaDepartment of Biomedical Engineering, University of VirginiaDepartment of Biomedical Engineering, University of VirginiaDepartment of Biomedical Engineering, University of VirginiaDepartment of Biomedical Engineering, University of VirginiaDepartment of Medicine, University of Virginia Health SystemDepartment of Cardiology, St. Francis HospitalCardiovascular Magnetic Resonance Unit, The Royal Brompton HospitalDepartment of Radiology and Imaging Sciences, Emory University School of MedicineDepartment of Radiology, Stanford UniversityDepartment of Biomedical Engineering, University of VirginiaAbstract Background Cardiovascular magnetic resonance (CMR) cine displacement encoding with stimulated echoes (DENSE) measures heart motion by encoding myocardial displacement into the signal phase, facilitating high accuracy and reproducibility of global and segmental myocardial strain and providing benefits in clinical performance. While conventional methods for strain analysis of DENSE images are faster than those for myocardial tagging, they still require manual user assistance. The present study developed and evaluated deep learning methods for fully-automatic DENSE strain analysis. Methods Convolutional neural networks (CNNs) were developed and trained to (a) identify the left-ventricular (LV) epicardial and endocardial borders, (b) identify the anterior right-ventricular (RV)-LV insertion point, and (c) perform phase unwrapping. Subsequent conventional automatic steps were employed to compute strain. The networks were trained using 12,415 short-axis DENSE images from 45 healthy subjects and 19 heart disease patients and were tested using 10,510 images from 25 healthy subjects and 19 patients. Each individual CNN was evaluated, and the end-to-end fully-automatic deep learning pipeline was compared to conventional user-assisted DENSE analysis using linear correlation and Bland Altman analysis of circumferential strain. Results LV myocardial segmentation U-Nets achieved a DICE similarity coefficient of 0.87 ± 0.04, a Hausdorff distance of 2.7 ± 1.0 pixels, and a mean surface distance of 0.41 ± 0.29 pixels in comparison with manual LV myocardial segmentation by an expert. The anterior RV-LV insertion point was detected within 1.38 ± 0.9 pixels compared to manually annotated data. The phase-unwrapping U-Net had similar or lower mean squared error vs. ground-truth data compared to the conventional path-following method for images with typical signal-to-noise ratio (SNR) or low SNR (p < 0.05), respectively. Bland–Altman analyses showed biases of 0.00 ± 0.03 and limits of agreement of − 0.04 to 0.05 or better for deep learning-based fully-automatic global and segmental end-systolic circumferential strain vs. conventional user-assisted methods. Conclusions Deep learning enables fully-automatic global and segmental circumferential strain analysis of DENSE CMR providing excellent agreement with conventional user-assisted methods. Deep learning-based automatic strain analysis may facilitate greater clinical use of DENSE for the quantification of global and segmental strain in patients with cardiac disease.https://doi.org/10.1186/s12968-021-00712-9DENSECardiac MRIMachine learningDeep learningPhase unwrappingStrain analysis |
| spellingShingle | Sona Ghadimi Daniel A. Auger Xue Feng Changyu Sun Craig H. Meyer Kenneth C. Bilchick Jie Jane Cao Andrew D. Scott John N. Oshinski Daniel B. Ennis Frederick H. Epstein Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping Journal of Cardiovascular Magnetic Resonance DENSE Cardiac MRI Machine learning Deep learning Phase unwrapping Strain analysis |
| title | Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping |
| title_full | Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping |
| title_fullStr | Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping |
| title_full_unstemmed | Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping |
| title_short | Fully-automated global and segmental strain analysis of DENSE cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping |
| title_sort | fully automated global and segmental strain analysis of dense cardiovascular magnetic resonance using deep learning for segmentation and phase unwrapping |
| topic | DENSE Cardiac MRI Machine learning Deep learning Phase unwrapping Strain analysis |
| url | https://doi.org/10.1186/s12968-021-00712-9 |
| work_keys_str_mv | AT sonaghadimi fullyautomatedglobalandsegmentalstrainanalysisofdensecardiovascularmagneticresonanceusingdeeplearningforsegmentationandphaseunwrapping AT danielaauger fullyautomatedglobalandsegmentalstrainanalysisofdensecardiovascularmagneticresonanceusingdeeplearningforsegmentationandphaseunwrapping AT xuefeng fullyautomatedglobalandsegmentalstrainanalysisofdensecardiovascularmagneticresonanceusingdeeplearningforsegmentationandphaseunwrapping AT changyusun fullyautomatedglobalandsegmentalstrainanalysisofdensecardiovascularmagneticresonanceusingdeeplearningforsegmentationandphaseunwrapping AT craighmeyer fullyautomatedglobalandsegmentalstrainanalysisofdensecardiovascularmagneticresonanceusingdeeplearningforsegmentationandphaseunwrapping AT kennethcbilchick fullyautomatedglobalandsegmentalstrainanalysisofdensecardiovascularmagneticresonanceusingdeeplearningforsegmentationandphaseunwrapping AT jiejanecao fullyautomatedglobalandsegmentalstrainanalysisofdensecardiovascularmagneticresonanceusingdeeplearningforsegmentationandphaseunwrapping AT andrewdscott fullyautomatedglobalandsegmentalstrainanalysisofdensecardiovascularmagneticresonanceusingdeeplearningforsegmentationandphaseunwrapping AT johnnoshinski fullyautomatedglobalandsegmentalstrainanalysisofdensecardiovascularmagneticresonanceusingdeeplearningforsegmentationandphaseunwrapping AT danielbennis fullyautomatedglobalandsegmentalstrainanalysisofdensecardiovascularmagneticresonanceusingdeeplearningforsegmentationandphaseunwrapping AT frederickhepstein fullyautomatedglobalandsegmentalstrainanalysisofdensecardiovascularmagneticresonanceusingdeeplearningforsegmentationandphaseunwrapping |