Explainable Anatomical Shape Analysis through Deep Hierarchical Generative Models
- Submitting institution
-
City, University of London
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 831
- Type
- D - Journal article
- DOI
-
10.1109/TMI.2020.2964499
- Title of journal
- IEEE Transactions on Medical Imaging
- Article number
- -
- First page
- 2088
- Volume
- 39
- Issue
- 6
- ISSN
- 0278-0062
- Open access status
- Compliant
- Month of publication
- January
- Year of publication
- 2020
- 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
- No
- Number of additional authors
-
14
- Research group(s)
-
-
- Citation count
- 2
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper explores for the first time a type of AI algorithm (called Ladder Variational Auto-Encoder) to analyse and interpret the 3D structure of organs (e.g. left ventricle, hippocampus) estimated from MRI. This approach paves the way for AI-based and data-driven understanding of structural changes in organs due to pathology. The study was tested and validated on two large datasets (cardiac and brain MRI respectively). It was supported by British Heart Foundation (BHF, NH/17/1/32725, RE/13/4/30184), Academy of Medical Sciences (SGL015/1006) and the NIHR Biomedical Research Centre. It was included as scientific basis for two grant applications to the BHF.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -
