Automated Endocardial Border Detection and Left Ventricular Functional Assessment in Echocardiography Using Deep Learning
Endocardial border detection is a key step in assessing left ventricular systolic function in echocardiography. However, this process is still not sufficiently accurate, and manual retracing is often required, causing time-consuming and intra-/inter-observer variability in clinical practice. To addr...
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-05-01
|
| Series: | Biomedicines |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2227-9059/10/5/1082 |
| _version_ | 1827670147838509056 |
|---|---|
| author | Shunzaburo Ono Masaaki Komatsu Akira Sakai Hideki Arima Mie Ochida Rina Aoyama Suguru Yasutomi Ken Asada Syuzo Kaneko Tetsuo Sasano Ryuji Hamamoto |
| author_facet | Shunzaburo Ono Masaaki Komatsu Akira Sakai Hideki Arima Mie Ochida Rina Aoyama Suguru Yasutomi Ken Asada Syuzo Kaneko Tetsuo Sasano Ryuji Hamamoto |
| author_sort | Shunzaburo Ono |
| collection | DOAJ |
| description | Endocardial border detection is a key step in assessing left ventricular systolic function in echocardiography. However, this process is still not sufficiently accurate, and manual retracing is often required, causing time-consuming and intra-/inter-observer variability in clinical practice. To address these clinical issues, more accurate and normalized automatic endocardial border detection would be valuable. Here, we develop a deep learning-based method for automated endocardial border detection and left ventricular functional assessment in two-dimensional echocardiographic videos. First, segmentation of the left ventricular cavity was performed in the six representative projections for a cardiac cycle. We employed four segmentation methods: U-Net, UNet++, UNet3+, and Deep Residual U-Net. UNet++ and UNet3+ showed a sufficiently high performance in the mean value of intersection over union and Dice coefficient. The accuracy of the four segmentation methods was then evaluated by calculating the mean value for the estimation error of the echocardiographic indexes. UNet++ was superior to the other segmentation methods, with the acceptable mean estimation error of the left ventricular ejection fraction of 10.8%, global longitudinal strain of 8.5%, and global circumferential strain of 5.8%, respectively. Our method using UNet++ demonstrated the best performance. This method may potentially support examiners and improve the workflow in echocardiography. |
| first_indexed | 2024-03-10T03:17:59Z |
| format | Article |
| id | doaj.art-e60b290eebe24b36b96403309299e742 |
| institution | Directory Open Access Journal |
| issn | 2227-9059 |
| language | English |
| last_indexed | 2024-03-10T03:17:59Z |
| publishDate | 2022-05-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Biomedicines |
| spelling | doaj.art-e60b290eebe24b36b96403309299e7422023-11-23T10:10:46ZengMDPI AGBiomedicines2227-90592022-05-01105108210.3390/biomedicines10051082Automated Endocardial Border Detection and Left Ventricular Functional Assessment in Echocardiography Using Deep LearningShunzaburo Ono0Masaaki Komatsu1Akira Sakai2Hideki Arima3Mie Ochida4Rina Aoyama5Suguru Yasutomi6Ken Asada7Syuzo Kaneko8Tetsuo Sasano9Ryuji Hamamoto10Department of Cardiovascular Medicine, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8510, JapanCancer Translational Research Team, RIKEN Center for Advanced Intelligence Project, 1-4-1 Nihonbashi, Chuo-ku, Tokyo 103-0027, JapanArtificial Intelligence Laboratory, Research Unit, Fujitsu Research, Fujitsu Ltd., 4-1-1 Kamikodanaka, Nakahara-ku, Kawasaki 211-8588, JapanDepartment of Cardiovascular Medicine, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8510, JapanDepartment of Cardiovascular Medicine, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8510, JapanDivision of Medical AI Research and Development, National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, JapanArtificial Intelligence Laboratory, Research Unit, Fujitsu Research, Fujitsu Ltd., 4-1-1 Kamikodanaka, Nakahara-ku, Kawasaki 211-8588, JapanDivision of Medical AI Research and Development, National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, JapanDivision of Medical AI Research and Development, National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, JapanDepartment of Cardiovascular Medicine, Tokyo Medical and Dental University, 1-5-45 Yushima, Bunkyo-ku, Tokyo 113-8510, JapanDivision of Medical AI Research and Development, National Cancer Center Research Institute, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, JapanEndocardial border detection is a key step in assessing left ventricular systolic function in echocardiography. However, this process is still not sufficiently accurate, and manual retracing is often required, causing time-consuming and intra-/inter-observer variability in clinical practice. To address these clinical issues, more accurate and normalized automatic endocardial border detection would be valuable. Here, we develop a deep learning-based method for automated endocardial border detection and left ventricular functional assessment in two-dimensional echocardiographic videos. First, segmentation of the left ventricular cavity was performed in the six representative projections for a cardiac cycle. We employed four segmentation methods: U-Net, UNet++, UNet3+, and Deep Residual U-Net. UNet++ and UNet3+ showed a sufficiently high performance in the mean value of intersection over union and Dice coefficient. The accuracy of the four segmentation methods was then evaluated by calculating the mean value for the estimation error of the echocardiographic indexes. UNet++ was superior to the other segmentation methods, with the acceptable mean estimation error of the left ventricular ejection fraction of 10.8%, global longitudinal strain of 8.5%, and global circumferential strain of 5.8%, respectively. Our method using UNet++ demonstrated the best performance. This method may potentially support examiners and improve the workflow in echocardiography.https://www.mdpi.com/2227-9059/10/5/1082deep learningechocardiographyendocardial border detectionleft ventricular ejection fractionmyocardial strain assessment |
| spellingShingle | Shunzaburo Ono Masaaki Komatsu Akira Sakai Hideki Arima Mie Ochida Rina Aoyama Suguru Yasutomi Ken Asada Syuzo Kaneko Tetsuo Sasano Ryuji Hamamoto Automated Endocardial Border Detection and Left Ventricular Functional Assessment in Echocardiography Using Deep Learning Biomedicines deep learning echocardiography endocardial border detection left ventricular ejection fraction myocardial strain assessment |
| title | Automated Endocardial Border Detection and Left Ventricular Functional Assessment in Echocardiography Using Deep Learning |
| title_full | Automated Endocardial Border Detection and Left Ventricular Functional Assessment in Echocardiography Using Deep Learning |
| title_fullStr | Automated Endocardial Border Detection and Left Ventricular Functional Assessment in Echocardiography Using Deep Learning |
| title_full_unstemmed | Automated Endocardial Border Detection and Left Ventricular Functional Assessment in Echocardiography Using Deep Learning |
| title_short | Automated Endocardial Border Detection and Left Ventricular Functional Assessment in Echocardiography Using Deep Learning |
| title_sort | automated endocardial border detection and left ventricular functional assessment in echocardiography using deep learning |
| topic | deep learning echocardiography endocardial border detection left ventricular ejection fraction myocardial strain assessment |
| url | https://www.mdpi.com/2227-9059/10/5/1082 |
| work_keys_str_mv | AT shunzaburoono automatedendocardialborderdetectionandleftventricularfunctionalassessmentinechocardiographyusingdeeplearning AT masaakikomatsu automatedendocardialborderdetectionandleftventricularfunctionalassessmentinechocardiographyusingdeeplearning AT akirasakai automatedendocardialborderdetectionandleftventricularfunctionalassessmentinechocardiographyusingdeeplearning AT hidekiarima automatedendocardialborderdetectionandleftventricularfunctionalassessmentinechocardiographyusingdeeplearning AT mieochida automatedendocardialborderdetectionandleftventricularfunctionalassessmentinechocardiographyusingdeeplearning AT rinaaoyama automatedendocardialborderdetectionandleftventricularfunctionalassessmentinechocardiographyusingdeeplearning AT suguruyasutomi automatedendocardialborderdetectionandleftventricularfunctionalassessmentinechocardiographyusingdeeplearning AT kenasada automatedendocardialborderdetectionandleftventricularfunctionalassessmentinechocardiographyusingdeeplearning AT syuzokaneko automatedendocardialborderdetectionandleftventricularfunctionalassessmentinechocardiographyusingdeeplearning AT tetsuosasano automatedendocardialborderdetectionandleftventricularfunctionalassessmentinechocardiographyusingdeeplearning AT ryujihamamoto automatedendocardialborderdetectionandleftventricularfunctionalassessmentinechocardiographyusingdeeplearning |