Leveraging Deep Learning for Approaching Automated Pre-Clinical Rodent Models
(2024) 16th International Joint Conference on Computational Intelligence, IJCCI 2024 p.613-620- Abstract
We evaluate deep learning architectures for rat pose estimation using a six-camera system, focusing on ResNet and EfficientNet across various depths and augmentation techniques. Among the configurations tested, ResNet 152 with default augmentation provided the best performance when employing a multi-perspective network approach in the controlled experimental setup. It reached a Root Mean Squared Error (RMSE) of 8.74, 8.78, and 9.72 pixels for the different angles. The utilization of data augmentation revealed that less altering yields better performance. We propose potential areas for future research, including further refinement of model configurations, more in-depth investigation of inference speeds, and the possibility of... (More)
We evaluate deep learning architectures for rat pose estimation using a six-camera system, focusing on ResNet and EfficientNet across various depths and augmentation techniques. Among the configurations tested, ResNet 152 with default augmentation provided the best performance when employing a multi-perspective network approach in the controlled experimental setup. It reached a Root Mean Squared Error (RMSE) of 8.74, 8.78, and 9.72 pixels for the different angles. The utilization of data augmentation revealed that less altering yields better performance. We propose potential areas for future research, including further refinement of model configurations, more in-depth investigation of inference speeds, and the possibility of transferring network weights to study other species, such as mice. The findings underscore the potential for deep learning solutions to advance preclinical research in behavioral neuroscience. We suggest building on this research to introduce behavioral recognition based on a 3D movement reconstruction, particularly emphasizing the motoric aspects of neurodegenerative diseases. This will allow for the correlation of observable behaviors with neuronal activity, contributing to a better understanding of the brain and aiding in developing new therapeutic strategies.
(Less)
- author
- Sandelius, Carl LU ; Pappas, Athanasios LU ; Sarkheyli-H Ä Gele, Arezoo ; Heuer, Andreas LU and Johnsson, Magnus
- organization
- publishing date
- 2024
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- keywords
- Behavioral Neuroscience, Computer Vision, Deep Learning, Machine Learning, Pre-Clinical Rodent Models
- host publication
- Proceedings of the 16th International Joint Conference on Computational Intelligence, IJCCI 2024
- editor
- Marcelloni, Francesco ; Madani, Kurosh ; van Stein, Niki and Joaquim, Joaquim
- pages
- 8 pages
- publisher
- Science and Technology Publications, Lda
- conference name
- 16th International Joint Conference on Computational Intelligence, IJCCI 2024
- conference location
- Porto, Portugal
- conference dates
- 2024-11-20 - 2024-11-22
- external identifiers
-
- scopus:85211432771
- ISBN
- 9789897587214
- DOI
- 10.5220/0013065600003837
- language
- English
- LU publication?
- yes
- id
- 44751214-c831-4b76-a96c-0af36d18568a
- date added to LUP
- 2025-01-27 13:12:48
- date last changed
- 2025-04-04 04:38:43
@inproceedings{44751214-c831-4b76-a96c-0af36d18568a,
abstract = {{<p>We evaluate deep learning architectures for rat pose estimation using a six-camera system, focusing on ResNet and EfficientNet across various depths and augmentation techniques. Among the configurations tested, ResNet 152 with default augmentation provided the best performance when employing a multi-perspective network approach in the controlled experimental setup. It reached a Root Mean Squared Error (RMSE) of 8.74, 8.78, and 9.72 pixels for the different angles. The utilization of data augmentation revealed that less altering yields better performance. We propose potential areas for future research, including further refinement of model configurations, more in-depth investigation of inference speeds, and the possibility of transferring network weights to study other species, such as mice. The findings underscore the potential for deep learning solutions to advance preclinical research in behavioral neuroscience. We suggest building on this research to introduce behavioral recognition based on a 3D movement reconstruction, particularly emphasizing the motoric aspects of neurodegenerative diseases. This will allow for the correlation of observable behaviors with neuronal activity, contributing to a better understanding of the brain and aiding in developing new therapeutic strategies.</p>}},
author = {{Sandelius, Carl and Pappas, Athanasios and Sarkheyli-H Ä Gele, Arezoo and Heuer, Andreas and Johnsson, Magnus}},
booktitle = {{Proceedings of the 16th International Joint Conference on Computational Intelligence, IJCCI 2024}},
editor = {{Marcelloni, Francesco and Madani, Kurosh and van Stein, Niki and Joaquim, Joaquim}},
isbn = {{9789897587214}},
keywords = {{Behavioral Neuroscience; Computer Vision; Deep Learning; Machine Learning; Pre-Clinical Rodent Models}},
language = {{eng}},
pages = {{613--620}},
publisher = {{Science and Technology Publications, Lda}},
title = {{Leveraging Deep Learning for Approaching Automated Pre-Clinical Rodent Models}},
url = {{http://dx.doi.org/10.5220/0013065600003837}},
doi = {{10.5220/0013065600003837}},
year = {{2024}},
}