Deep learning for automated left ventricular outflow tract diameter measurements in 2D echocardiography
Abstract Background Measurement of the left ventricular outflow tract diameter (LVOTd) in echocardiography is a common source of error when used to calculate the stroke volume. The aim of this study is to assess whether a deep learning (DL) model, trained on a clinical echocardiographic dataset, can...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2023-10-01
|
| Series: | Cardiovascular Ultrasound |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12947-023-00317-5 |
| _version_ | 1797577843727663104 |
|---|---|
| author | Sigurd Zijun Zha Magnus Rogstadkjernet Lars Gunnar Klæboe Helge Skulstad Bjørn-Jostein Singstad Andrew Gilbert Thor Edvardsen Eigil Samset Pål Haugar Brekke |
| author_facet | Sigurd Zijun Zha Magnus Rogstadkjernet Lars Gunnar Klæboe Helge Skulstad Bjørn-Jostein Singstad Andrew Gilbert Thor Edvardsen Eigil Samset Pål Haugar Brekke |
| author_sort | Sigurd Zijun Zha |
| collection | DOAJ |
| description | Abstract Background Measurement of the left ventricular outflow tract diameter (LVOTd) in echocardiography is a common source of error when used to calculate the stroke volume. The aim of this study is to assess whether a deep learning (DL) model, trained on a clinical echocardiographic dataset, can perform automatic LVOTd measurements on par with expert cardiologists. Methods Data consisted of 649 consecutive transthoracic echocardiographic examinations of patients with coronary artery disease admitted to a university hospital. 1304 LVOTd measurements in the parasternal long axis (PLAX) and zoomed parasternal long axis views (ZPLAX) were collected, with each patient having 1–6 measurements per examination. Data quality control was performed by an expert cardiologist, and spatial geometry data was preserved for each LVOTd measurement to convert DL predictions into metric units. A convolutional neural network based on the U-Net was used as the DL model. Results The mean absolute LVOTd error was 1.04 (95% confidence interval [CI] 0.90–1.19) mm for DL predictions on the test set. The mean relative LVOTd errors across all data subgroups ranged from 3.8 to 5.1% for the test set. Generally, the DL model had superior performance on the ZPLAX view compared to the PLAX view. DL model precision for patients with repeated LVOTd measurements had a mean coefficient of variation of 2.2 (95% CI 1.6–2.7) %, which was comparable to the clinicians for the test set. Conclusion DL for automatic LVOTd measurements in PLAX and ZPLAX views is feasible when trained on a limited clinical dataset. While the DL predicted LVOTd measurements were within the expected range of clinical inter-observer variability, the robustness of the DL model requires validation on independent datasets. Future experiments using temporal information and anatomical constraints could improve valvular identification and reduce outliers, which are challenges that must be addressed before clinical utilization. Graphical Abstract |
| first_indexed | 2024-03-10T22:14:41Z |
| format | Article |
| id | doaj.art-c992a1a010894ccd9245821d60352255 |
| institution | Directory Open Access Journal |
| issn | 1476-7120 |
| language | English |
| last_indexed | 2024-03-10T22:14:41Z |
| publishDate | 2023-10-01 |
| publisher | BMC |
| record_format | Article |
| series | Cardiovascular Ultrasound |
| spelling | doaj.art-c992a1a010894ccd9245821d603522552023-11-19T12:30:37ZengBMCCardiovascular Ultrasound1476-71202023-10-0121111110.1186/s12947-023-00317-5Deep learning for automated left ventricular outflow tract diameter measurements in 2D echocardiographySigurd Zijun Zha0Magnus Rogstadkjernet1Lars Gunnar Klæboe2Helge Skulstad3Bjørn-Jostein Singstad4Andrew Gilbert5Thor Edvardsen6Eigil Samset7Pål Haugar Brekke8University of OsloUniversity of OsloOslo University Hospital, RikshospitaletUniversity of OsloOslo University Hospital, RikshospitaletGE HealthCareUniversity of OsloUniversity of OsloOslo University Hospital, RikshospitaletAbstract Background Measurement of the left ventricular outflow tract diameter (LVOTd) in echocardiography is a common source of error when used to calculate the stroke volume. The aim of this study is to assess whether a deep learning (DL) model, trained on a clinical echocardiographic dataset, can perform automatic LVOTd measurements on par with expert cardiologists. Methods Data consisted of 649 consecutive transthoracic echocardiographic examinations of patients with coronary artery disease admitted to a university hospital. 1304 LVOTd measurements in the parasternal long axis (PLAX) and zoomed parasternal long axis views (ZPLAX) were collected, with each patient having 1–6 measurements per examination. Data quality control was performed by an expert cardiologist, and spatial geometry data was preserved for each LVOTd measurement to convert DL predictions into metric units. A convolutional neural network based on the U-Net was used as the DL model. Results The mean absolute LVOTd error was 1.04 (95% confidence interval [CI] 0.90–1.19) mm for DL predictions on the test set. The mean relative LVOTd errors across all data subgroups ranged from 3.8 to 5.1% for the test set. Generally, the DL model had superior performance on the ZPLAX view compared to the PLAX view. DL model precision for patients with repeated LVOTd measurements had a mean coefficient of variation of 2.2 (95% CI 1.6–2.7) %, which was comparable to the clinicians for the test set. Conclusion DL for automatic LVOTd measurements in PLAX and ZPLAX views is feasible when trained on a limited clinical dataset. While the DL predicted LVOTd measurements were within the expected range of clinical inter-observer variability, the robustness of the DL model requires validation on independent datasets. Future experiments using temporal information and anatomical constraints could improve valvular identification and reduce outliers, which are challenges that must be addressed before clinical utilization. Graphical Abstracthttps://doi.org/10.1186/s12947-023-00317-5Deep learningLeft ventricular outflow tractTransthoracic echocardiographyMachine learningAutomated measurements |
| spellingShingle | Sigurd Zijun Zha Magnus Rogstadkjernet Lars Gunnar Klæboe Helge Skulstad Bjørn-Jostein Singstad Andrew Gilbert Thor Edvardsen Eigil Samset Pål Haugar Brekke Deep learning for automated left ventricular outflow tract diameter measurements in 2D echocardiography Cardiovascular Ultrasound Deep learning Left ventricular outflow tract Transthoracic echocardiography Machine learning Automated measurements |
| title | Deep learning for automated left ventricular outflow tract diameter measurements in 2D echocardiography |
| title_full | Deep learning for automated left ventricular outflow tract diameter measurements in 2D echocardiography |
| title_fullStr | Deep learning for automated left ventricular outflow tract diameter measurements in 2D echocardiography |
| title_full_unstemmed | Deep learning for automated left ventricular outflow tract diameter measurements in 2D echocardiography |
| title_short | Deep learning for automated left ventricular outflow tract diameter measurements in 2D echocardiography |
| title_sort | deep learning for automated left ventricular outflow tract diameter measurements in 2d echocardiography |
| topic | Deep learning Left ventricular outflow tract Transthoracic echocardiography Machine learning Automated measurements |
| url | https://doi.org/10.1186/s12947-023-00317-5 |
| work_keys_str_mv | AT sigurdzijunzha deeplearningforautomatedleftventricularoutflowtractdiametermeasurementsin2dechocardiography AT magnusrogstadkjernet deeplearningforautomatedleftventricularoutflowtractdiametermeasurementsin2dechocardiography AT larsgunnarklæboe deeplearningforautomatedleftventricularoutflowtractdiametermeasurementsin2dechocardiography AT helgeskulstad deeplearningforautomatedleftventricularoutflowtractdiametermeasurementsin2dechocardiography AT bjørnjosteinsingstad deeplearningforautomatedleftventricularoutflowtractdiametermeasurementsin2dechocardiography AT andrewgilbert deeplearningforautomatedleftventricularoutflowtractdiametermeasurementsin2dechocardiography AT thoredvardsen deeplearningforautomatedleftventricularoutflowtractdiametermeasurementsin2dechocardiography AT eigilsamset deeplearningforautomatedleftventricularoutflowtractdiametermeasurementsin2dechocardiography AT palhaugarbrekke deeplearningforautomatedleftventricularoutflowtractdiametermeasurementsin2dechocardiography |