Automatic Aortic Valve Extraction Using Deep Learning with Contrast-Enhanced Cardiac CT Images
Purpose: This study evaluates the use of deep learning techniques to automatically extract and delineate the aortic valve annulus region from contrast-enhanced cardiac CT images. Two approaches, namely, segmentation and object detection, were compared to determine their accuracy. Materials and Metho...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Journal of Cardiovascular Development and Disease |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2308-3425/12/1/3 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832588288591921152 |
|---|---|
| author | Soichiro Inomata Takaaki Yoshimura Minghui Tang Shota Ichikawa Hiroyuki Sugimori |
| author_facet | Soichiro Inomata Takaaki Yoshimura Minghui Tang Shota Ichikawa Hiroyuki Sugimori |
| author_sort | Soichiro Inomata |
| collection | DOAJ |
| description | Purpose: This study evaluates the use of deep learning techniques to automatically extract and delineate the aortic valve annulus region from contrast-enhanced cardiac CT images. Two approaches, namely, segmentation and object detection, were compared to determine their accuracy. Materials and Methods: A dataset of 32 contrast-enhanced cardiac CT scans was analyzed. The segmentation approach utilized the DeepLabv3+ model, while the object detection approach employed YOLOv2. The dataset was augmented through rotation and scaling, and five-fold cross-validation was applied. The accuracy of both methods was evaluated using the Dice similarity coefficient (DSC), and their performance in estimating the aortic valve annulus area was compared. Results: The object detection approach achieved a mean DSC of 0.809, significantly outperforming the segmentation approach, which had a mean DSC of 0.711. Object detection also demonstrated higher precision and recall, with fewer false positives and negatives. The aortic valve annulus area estimation had a mean error of 2.55 mm. Conclusions: Object detection showed superior performance in identifying the aortic valve annulus region, suggesting its potential for clinical application in cardiac imaging. The results highlight the promise of deep learning in improving the accuracy and efficiency of preoperative planning for cardiovascular interventions. |
| format | Article |
| id | doaj-art-8e850d6fc7474c58976ad86f3af0c24d |
| institution | Kabale University |
| issn | 2308-3425 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Journal of Cardiovascular Development and Disease |
| spelling | doaj-art-8e850d6fc7474c58976ad86f3af0c24d2025-01-24T13:35:56ZengMDPI AGJournal of Cardiovascular Development and Disease2308-34252024-12-01121310.3390/jcdd12010003Automatic Aortic Valve Extraction Using Deep Learning with Contrast-Enhanced Cardiac CT ImagesSoichiro Inomata0Takaaki Yoshimura1Minghui Tang2Shota Ichikawa3Hiroyuki Sugimori4Graduate School of Health Sciences, Hokkaido University, Sapporo 060-0812, JapanDepartment of Health Sciences and Technology, Faculty of Health Sciences, Hokkaido University, Sapporo 060-0812, JapanClinical AI Human Resources Development Program, Faculty of Medicine, Hokkaido University, Sapporo 060-8648, JapanDepartment of Radiological Technology, School of Health Sciences, Faculty of Medicine, Niigata University, Niigata 951-8518, JapanGlobal Center for Biomedical Science and Engineering, Faculty of Medicine, Hokkaido University, Sapporo 060-8638, JapanPurpose: This study evaluates the use of deep learning techniques to automatically extract and delineate the aortic valve annulus region from contrast-enhanced cardiac CT images. Two approaches, namely, segmentation and object detection, were compared to determine their accuracy. Materials and Methods: A dataset of 32 contrast-enhanced cardiac CT scans was analyzed. The segmentation approach utilized the DeepLabv3+ model, while the object detection approach employed YOLOv2. The dataset was augmented through rotation and scaling, and five-fold cross-validation was applied. The accuracy of both methods was evaluated using the Dice similarity coefficient (DSC), and their performance in estimating the aortic valve annulus area was compared. Results: The object detection approach achieved a mean DSC of 0.809, significantly outperforming the segmentation approach, which had a mean DSC of 0.711. Object detection also demonstrated higher precision and recall, with fewer false positives and negatives. The aortic valve annulus area estimation had a mean error of 2.55 mm. Conclusions: Object detection showed superior performance in identifying the aortic valve annulus region, suggesting its potential for clinical application in cardiac imaging. The results highlight the promise of deep learning in improving the accuracy and efficiency of preoperative planning for cardiovascular interventions.https://www.mdpi.com/2308-3425/12/1/3deep learningconvolution neural networkaortic valve |
| spellingShingle | Soichiro Inomata Takaaki Yoshimura Minghui Tang Shota Ichikawa Hiroyuki Sugimori Automatic Aortic Valve Extraction Using Deep Learning with Contrast-Enhanced Cardiac CT Images Journal of Cardiovascular Development and Disease deep learning convolution neural network aortic valve |
| title | Automatic Aortic Valve Extraction Using Deep Learning with Contrast-Enhanced Cardiac CT Images |
| title_full | Automatic Aortic Valve Extraction Using Deep Learning with Contrast-Enhanced Cardiac CT Images |
| title_fullStr | Automatic Aortic Valve Extraction Using Deep Learning with Contrast-Enhanced Cardiac CT Images |
| title_full_unstemmed | Automatic Aortic Valve Extraction Using Deep Learning with Contrast-Enhanced Cardiac CT Images |
| title_short | Automatic Aortic Valve Extraction Using Deep Learning with Contrast-Enhanced Cardiac CT Images |
| title_sort | automatic aortic valve extraction using deep learning with contrast enhanced cardiac ct images |
| topic | deep learning convolution neural network aortic valve |
| url | https://www.mdpi.com/2308-3425/12/1/3 |
| work_keys_str_mv | AT soichiroinomata automaticaorticvalveextractionusingdeeplearningwithcontrastenhancedcardiacctimages AT takaakiyoshimura automaticaorticvalveextractionusingdeeplearningwithcontrastenhancedcardiacctimages AT minghuitang automaticaorticvalveextractionusingdeeplearningwithcontrastenhancedcardiacctimages AT shotaichikawa automaticaorticvalveextractionusingdeeplearningwithcontrastenhancedcardiacctimages AT hiroyukisugimori automaticaorticvalveextractionusingdeeplearningwithcontrastenhancedcardiacctimages |