Conference proceeding
Multi-agent reinforcement learning pipeline for anatomical landmark detection in minipigs
Vol.12032, pp.1203229-1203229-10
04/04/2022
DOI: 10.1117/12.2611008
Abstract
In recent years, the use of large animals in neurological research has escalated due to advantages over small animals. Unfortunately, large animal imaging researchers lack functional automated medical imaging tools, requiring laborious manual processing. As a response, we have implemented a Reinforcement Learning pipeline for brain anatomical landmark detection in minipig MRIs. Leveraging a deep convolutional network, two-step detection process, and multiple Deep-Q multi-agent networks, our approach is suitable for accurate landmark detection in large animals. Using a heterogeneous dataset containing 154 minipig images, we achieved an average accuracy of 1.56mm on predicting 19 landmarks.
Details
- Title: Subtitle
- Multi-agent reinforcement learning pipeline for anatomical landmark detection in minipigs
- Creators
- Michal Brzus - University of IowaAlexander B. Powers - The Univ. of Iowa (United States)Kevin S. Knoernschild - The Univ. of Iowa (United States)Jessica C. Sieren - University of IowaHans J. Johnson - University of Iowa
- Contributors
- Olivier Colliot (Editor) - Ctr. National de la Recherche Scientifique (France)Ivana Išgum (Editor) - Amsterdam UMC (Netherlands)
- Resource Type
- Conference proceeding
- Publication Details
- Vol.12032, pp.1203229-1203229-10
- Publisher
- SPIE
- DOI
- 10.1117/12.2611008
- ISSN
- 1605-7422
- Language
- English
- Date published
- 04/04/2022
- Academic Unit
- Electrical and Computer Engineering; Roy J. Carver Department of Biomedical Engineering; Iowa Informatics Initiative; The Iowa Initiative for Artificial Intelligence; Radiology; Psychiatry; The Iowa Institute for Biomedical Imaging
- Record Identifier
- 9984259366002771
Metrics
17 Record Views