Automatic 3D bi-ventricular segmentation of cardiac images by a shape-refined multi-task deep learning approach
- Submitting institution
-
The University of Birmingham
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 82801015
- Type
- D - Journal article
- DOI
-
10.1109/TMI.2019.2894322
- Title of journal
- IEEE Transactions on Medical Imaging
- Article number
- -
- First page
- 2151
- Volume
- 38
- Issue
- 9
- ISSN
- 0278-0062
- Open access status
- Compliant
- Month of publication
- January
- Year of publication
- 2019
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- Yes
- Number of additional authors
-
9
- Research group(s)
-
-
- Citation count
- 30
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This work highlights the significance of integrating anatomical prior knowledge to a deep learning–based segmentation method to analyse bi-ventricular size and function in a large-scale cohort (1831 healthy subjects and 649 pulmonary hypertension (PH) patients). World-leading computational imaging researchers, such as Prof Alejandro Frangi (Leeds), Prof Frederik Maes (KU Leuven), Prof Pierre-Marc Jodin (Sherbrooke) and Prof Vicente Grau (Oxford), have taken up the work. A recent clinical research project conducted by the prestigious Feinberg Cardiovascular Research Institute commented on our work thus: "These refinements to existing deep learning approaches will likely be of significant value in future PH clinical trials."
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -