3D behavior phenotyping and multi-object tracking system of mice based on deep learning

To extract multiple-mice behaviors automatically, nondestructively, and long-termly, a 3D behavior extraction system was developed for extracting individual behaviors and social behaviors of mice. In this study, two depth cameras were used to acquire 3D information of mice. 3D point clouds model of...

Full description

Saved in:
Bibliographic Details
Main Authors: Lei He, Xuezhen Jia, Chen Li, Yan Liu, Jicheng Yu, Zhen-Xia Chen, Wanneng Yang, Xiuying Liang
Format: Article
Language:English
Published: Elsevier 2025-08-01
Series:Smart Agricultural Technology
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2772375525002552
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849326050999271424
author Lei He
Xuezhen Jia
Chen Li
Yan Liu
Jicheng Yu
Zhen-Xia Chen
Wanneng Yang
Xiuying Liang
author_facet Lei He
Xuezhen Jia
Chen Li
Yan Liu
Jicheng Yu
Zhen-Xia Chen
Wanneng Yang
Xiuying Liang
author_sort Lei He
collection DOAJ
description To extract multiple-mice behaviors automatically, nondestructively, and long-termly, a 3D behavior extraction system was developed for extracting individual behaviors and social behaviors of mice. In this study, two depth cameras were used to acquire 3D information of mice. 3D point clouds model of mice was established and the algorithms of clipping, statistical filter, matching, plane segmenting, and rotating were used to preprocess the 3D point clouds. In this study, YOLOv7 was used to detect mice and 3D Kalman Filter, Hungarian algorithm, and improved matching algorithm were used to track mouse identity. Stratified Transformer was used to segment individual mouse from the environment and segment individual mouse into 3 parts, and 8 feature points were extracted. In individual mouse experiment, the whole individual mouse was segmented to 4 parts and 11 feature points were extracted. Individual and social behaviors of mice were extracted combined feature points and mice trajectories. The results showed the accuracy and average mIoU of individual mouse segmentation were 95.80 % and 91.90 %, respectively, and the accuracy and average mIoU of parts segmentation were 93.60 % and 85.9 %, respectively. Compared with manual measurement of mice weight, the determination coefficient (R2) of mouse volume was 0.9726. The maximum absolute error of the automatic measurements versus manual measurements of mouse tail length was 0.4 mm. The precision, recall, and mAP of YOLOv7 were 99.76 %, 99.53 %, and 99.75 %, respectively. The analysis of behaviors also showed that there were significant differences between dexamethasone-induced muscle atrophy and normal mice.
format Article
id doaj-art-884c8e37f83c42689a75f2effccd263e
institution Kabale University
issn 2772-3755
language English
publishDate 2025-08-01
publisher Elsevier
record_format Article
series Smart Agricultural Technology
spelling doaj-art-884c8e37f83c42689a75f2effccd263e2025-08-20T03:48:14ZengElsevierSmart Agricultural Technology2772-37552025-08-011110102210.1016/j.atech.2025.1010223D behavior phenotyping and multi-object tracking system of mice based on deep learningLei He0Xuezhen Jia1Chen Li2Yan Liu3Jicheng Yu4Zhen-Xia Chen5Wanneng Yang6Xiuying Liang7National Key Laboratory of Crop Genetic Improvement, National Center of Plant Gene Research (Wuhan), Hubei Hongshan Laboratory, College of Engineering, College of Informatics, College of Animal Science & Technology, College of Biomedicine and Health, Huazhong Agricultural University, Wuhan 430070, Hubei, PR ChinaNational Key Laboratory of Crop Genetic Improvement, National Center of Plant Gene Research (Wuhan), Hubei Hongshan Laboratory, College of Engineering, College of Informatics, College of Animal Science & Technology, College of Biomedicine and Health, Huazhong Agricultural University, Wuhan 430070, Hubei, PR ChinaNational Key Laboratory of Crop Genetic Improvement, National Center of Plant Gene Research (Wuhan), Hubei Hongshan Laboratory, College of Engineering, College of Informatics, College of Animal Science & Technology, College of Biomedicine and Health, Huazhong Agricultural University, Wuhan 430070, Hubei, PR ChinaNational Key Laboratory of Crop Genetic Improvement, National Center of Plant Gene Research (Wuhan), Hubei Hongshan Laboratory, College of Engineering, College of Informatics, College of Animal Science & Technology, College of Biomedicine and Health, Huazhong Agricultural University, Wuhan 430070, Hubei, PR ChinaNational Key Laboratory of Crop Genetic Improvement, National Center of Plant Gene Research (Wuhan), Hubei Hongshan Laboratory, College of Engineering, College of Informatics, College of Animal Science & Technology, College of Biomedicine and Health, Huazhong Agricultural University, Wuhan 430070, Hubei, PR ChinaNational Key Laboratory of Crop Genetic Improvement, National Center of Plant Gene Research (Wuhan), Hubei Hongshan Laboratory, College of Engineering, College of Informatics, College of Animal Science & Technology, College of Biomedicine and Health, Huazhong Agricultural University, Wuhan 430070, Hubei, PR ChinaNational Key Laboratory of Crop Genetic Improvement, National Center of Plant Gene Research (Wuhan), Hubei Hongshan Laboratory, College of Engineering, College of Informatics, College of Animal Science & Technology, College of Biomedicine and Health, Huazhong Agricultural University, Wuhan 430070, Hubei, PR China; Shenzhen Branch, Guangdong Laboratory for Lingnan Modern Agriculture, Genome Analysis Laboratory of the Ministry of Agriculture, Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural Sciences, Shenzhen, PR China; Corresponding authors.National Key Laboratory of Crop Genetic Improvement, National Center of Plant Gene Research (Wuhan), Hubei Hongshan Laboratory, College of Engineering, College of Informatics, College of Animal Science & Technology, College of Biomedicine and Health, Huazhong Agricultural University, Wuhan 430070, Hubei, PR China; Corresponding authors.To extract multiple-mice behaviors automatically, nondestructively, and long-termly, a 3D behavior extraction system was developed for extracting individual behaviors and social behaviors of mice. In this study, two depth cameras were used to acquire 3D information of mice. 3D point clouds model of mice was established and the algorithms of clipping, statistical filter, matching, plane segmenting, and rotating were used to preprocess the 3D point clouds. In this study, YOLOv7 was used to detect mice and 3D Kalman Filter, Hungarian algorithm, and improved matching algorithm were used to track mouse identity. Stratified Transformer was used to segment individual mouse from the environment and segment individual mouse into 3 parts, and 8 feature points were extracted. In individual mouse experiment, the whole individual mouse was segmented to 4 parts and 11 feature points were extracted. Individual and social behaviors of mice were extracted combined feature points and mice trajectories. The results showed the accuracy and average mIoU of individual mouse segmentation were 95.80 % and 91.90 %, respectively, and the accuracy and average mIoU of parts segmentation were 93.60 % and 85.9 %, respectively. Compared with manual measurement of mice weight, the determination coefficient (R2) of mouse volume was 0.9726. The maximum absolute error of the automatic measurements versus manual measurements of mouse tail length was 0.4 mm. The precision, recall, and mAP of YOLOv7 were 99.76 %, 99.53 %, and 99.75 %, respectively. The analysis of behaviors also showed that there were significant differences between dexamethasone-induced muscle atrophy and normal mice.http://www.sciencedirect.com/science/article/pii/S2772375525002552Muscular atrophy miceIndividual/social behaviors3D behavior phenotyping systemMulti-object detectionMulti-object tracking
spellingShingle Lei He
Xuezhen Jia
Chen Li
Yan Liu
Jicheng Yu
Zhen-Xia Chen
Wanneng Yang
Xiuying Liang
3D behavior phenotyping and multi-object tracking system of mice based on deep learning
Smart Agricultural Technology
Muscular atrophy mice
Individual/social behaviors
3D behavior phenotyping system
Multi-object detection
Multi-object tracking
title 3D behavior phenotyping and multi-object tracking system of mice based on deep learning
title_full 3D behavior phenotyping and multi-object tracking system of mice based on deep learning
title_fullStr 3D behavior phenotyping and multi-object tracking system of mice based on deep learning
title_full_unstemmed 3D behavior phenotyping and multi-object tracking system of mice based on deep learning
title_short 3D behavior phenotyping and multi-object tracking system of mice based on deep learning
title_sort 3d behavior phenotyping and multi object tracking system of mice based on deep learning
topic Muscular atrophy mice
Individual/social behaviors
3D behavior phenotyping system
Multi-object detection
Multi-object tracking
url http://www.sciencedirect.com/science/article/pii/S2772375525002552
work_keys_str_mv AT leihe 3dbehaviorphenotypingandmultiobjecttrackingsystemofmicebasedondeeplearning
AT xuezhenjia 3dbehaviorphenotypingandmultiobjecttrackingsystemofmicebasedondeeplearning
AT chenli 3dbehaviorphenotypingandmultiobjecttrackingsystemofmicebasedondeeplearning
AT yanliu 3dbehaviorphenotypingandmultiobjecttrackingsystemofmicebasedondeeplearning
AT jichengyu 3dbehaviorphenotypingandmultiobjecttrackingsystemofmicebasedondeeplearning
AT zhenxiachen 3dbehaviorphenotypingandmultiobjecttrackingsystemofmicebasedondeeplearning
AT wannengyang 3dbehaviorphenotypingandmultiobjecttrackingsystemofmicebasedondeeplearning
AT xiuyingliang 3dbehaviorphenotypingandmultiobjecttrackingsystemofmicebasedondeeplearning