Anatomically Constrained Neural Networks (ACNNs): Application to Cardiac Image Enhancement and Segmentation
- Submitting institution
-
Imperial College of Science, Technology and Medicine
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 2145
- Type
- D - Journal article
- DOI
-
10.1109/TMI.2017.2743464
- Title of journal
- IEEE Transactions on Medical Imaging
- Article number
- 2
- First page
- 384
- Volume
- 37
- Issue
- 2
- ISSN
- 0278-0062
- Open access status
- Compliant
- Month of publication
- September
- Year of publication
- 2017
- URL
-
-
- Supplementary information
-
10.1109/TMI.2017.2743464
- 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
-
12
- Research group(s)
-
-
- Citation count
- 149
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- The first paper to propose the use of anatomical priors for convolutional neural networks in medical imaging, demonstrating applications for image segmentation and image super-resolution. The paper was instrumental in establishing an industrial collaboration with Heartflow (funding >£1M, www.heartflow.com). The paper is an extension of a highly-cited MICCAI 2016 paper (https://link.springer.com/chapter/10.1007/978-3-319-46726-9_29, oral presentation, acceptance rate: <4%). The work led to Oktay being offered a senior research position at Microsoft Research Cambridge.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
