Automated heart segmentation using U-Net in pediatric cardiac CT
This study investigated the usefulness of deep learning methods for segmenting the whole heart region and the cardiac cavity region in pediatric cardiac CT images using U-Net. Dice similarity coefficient (DSC) was used to evaluate the segmentation accuracy by leave-one-subject-out cross-validation....
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2021-12-01
|
| Series: | Measurement: Sensors |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2665917421000908 |
| _version_ | 1818992176481173504 |
|---|---|
| author | Akifumi Yoshida Yongbum Lee Norihiko Yoshimura Tatsuya Kuramoto Akira Hasegawa Tsutomu Kanazawa |
| author_facet | Akifumi Yoshida Yongbum Lee Norihiko Yoshimura Tatsuya Kuramoto Akira Hasegawa Tsutomu Kanazawa |
| author_sort | Akifumi Yoshida |
| collection | DOAJ |
| description | This study investigated the usefulness of deep learning methods for segmenting the whole heart region and the cardiac cavity region in pediatric cardiac CT images using U-Net. Dice similarity coefficient (DSC) was used to evaluate the segmentation accuracy by leave-one-subject-out cross-validation. The mean DSC for the whole heart was over 0.95, and analysis of variance among the four age categories (less than one year, 1y to 4y, 5y to 9y, 10y to 14y) showed no significant differences. The mean DSCs for each chamber were 0.78–0.88 when they were trained in a lump. The corresponding DSCs were 0.80–0.85 when they were trained separately. Although the size and shape of the heart varied with age in children, whole heart segmentation using U-Net showed high DSCs in all age categories. Deep learning would become a useful elemental technology in heart segmentation of pediatric cardiology. |
| first_indexed | 2024-12-20T20:22:00Z |
| format | Article |
| id | doaj.art-569589a60abe47b9ad8d5474cc406cf3 |
| institution | Directory Open Access Journal |
| issn | 2665-9174 |
| language | English |
| last_indexed | 2024-12-20T20:22:00Z |
| publishDate | 2021-12-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Measurement: Sensors |
| spelling | doaj.art-569589a60abe47b9ad8d5474cc406cf32022-12-21T19:27:33ZengElsevierMeasurement: Sensors2665-91742021-12-0118100127Automated heart segmentation using U-Net in pediatric cardiac CTAkifumi Yoshida0Yongbum Lee1Norihiko Yoshimura2Tatsuya Kuramoto3Akira Hasegawa4Tsutomu Kanazawa5Corresponding author.; Department of Radiological Technology, Faculty of Medical Technology, Niigata University of Health and Welfare, Niigata, Japan; Department of Radiological Technology, Graduate School of Health Science, Niigata University, Niigata, JapanDepartment of Radiological Technology, Graduate School of Health Science, Niigata University, Niigata, JapanDepartment of Radiology, Niigata City General Hospital, Niigata, JapanDepartment of Radiological Technology, Niigata University Medical & Dental Hospital, Niigata City, JapanDepartment of Radiological Technology, Faculty of Medical Technology, Niigata University of Health and Welfare, Niigata, JapanDepartment of Radiological Technology, Niigata University Medical & Dental Hospital, Niigata City, JapanThis study investigated the usefulness of deep learning methods for segmenting the whole heart region and the cardiac cavity region in pediatric cardiac CT images using U-Net. Dice similarity coefficient (DSC) was used to evaluate the segmentation accuracy by leave-one-subject-out cross-validation. The mean DSC for the whole heart was over 0.95, and analysis of variance among the four age categories (less than one year, 1y to 4y, 5y to 9y, 10y to 14y) showed no significant differences. The mean DSCs for each chamber were 0.78–0.88 when they were trained in a lump. The corresponding DSCs were 0.80–0.85 when they were trained separately. Although the size and shape of the heart varied with age in children, whole heart segmentation using U-Net showed high DSCs in all age categories. Deep learning would become a useful elemental technology in heart segmentation of pediatric cardiology.http://www.sciencedirect.com/science/article/pii/S2665917421000908SegmentationPediatricHeartComputed tomographyDeep learning |
| spellingShingle | Akifumi Yoshida Yongbum Lee Norihiko Yoshimura Tatsuya Kuramoto Akira Hasegawa Tsutomu Kanazawa Automated heart segmentation using U-Net in pediatric cardiac CT Measurement: Sensors Segmentation Pediatric Heart Computed tomography Deep learning |
| title | Automated heart segmentation using U-Net in pediatric cardiac CT |
| title_full | Automated heart segmentation using U-Net in pediatric cardiac CT |
| title_fullStr | Automated heart segmentation using U-Net in pediatric cardiac CT |
| title_full_unstemmed | Automated heart segmentation using U-Net in pediatric cardiac CT |
| title_short | Automated heart segmentation using U-Net in pediatric cardiac CT |
| title_sort | automated heart segmentation using u net in pediatric cardiac ct |
| topic | Segmentation Pediatric Heart Computed tomography Deep learning |
| url | http://www.sciencedirect.com/science/article/pii/S2665917421000908 |
| work_keys_str_mv | AT akifumiyoshida automatedheartsegmentationusingunetinpediatriccardiacct AT yongbumlee automatedheartsegmentationusingunetinpediatriccardiacct AT norihikoyoshimura automatedheartsegmentationusingunetinpediatriccardiacct AT tatsuyakuramoto automatedheartsegmentationusingunetinpediatriccardiacct AT akirahasegawa automatedheartsegmentationusingunetinpediatriccardiacct AT tsutomukanazawa automatedheartsegmentationusingunetinpediatriccardiacct |