Improving behavioral test data collection and analysis in animal models with an image processing program

Majid Davidson; Niloufar Rashidi; Puspha Sinnayah; Amir Hossein Ahmadi; Vasso Apostolopoulos; Kulmira Nurgali

doi:10.1016/j.bbr.2023.114544

Improving behavioral test data collection and analysis in animal models with an image processing program

Behav Brain Res. 2023 Aug 24:452:114544. doi: 10.1016/j.bbr.2023.114544. Epub 2023 Jun 14.

Authors

Majid Davidson¹, Niloufar Rashidi¹, Puspha Sinnayah¹, Amir Hossein Ahmadi², Vasso Apostolopoulos³, Kulmira Nurgali⁴

Affiliations

¹ Institute for Health and Sport, Victoria University, Melbourne, Australia.
² Department of Mechanical Engineering, K. N. Toosi University of Technology, Tehran, Iran.
³ Institute for Health and Sport, Victoria University, Melbourne, Australia; Immunology Program, Australian Institute of Musculoskeletal Science (AIMSS), Melbourne, Victoria, Australia. Electronic address: Vasso.Apostolopoulos@vu.edu.au.
⁴ Institute for Health and Sport, Victoria University, Melbourne, Australia; Department of Medicine Western Health, Faculty of Medicine, Dentistry and Health Sciences, University of Melbourne, Melbourne, Australia; Regenerative Medicine and Stem Cells Program, Australian Institute of Musculoskeletal Science (AIMSS), Melbourne, Victoria, Australia. Electronic address: Kulmira.Nurgali@vu.edu.au.

PMID: 37321312
DOI: 10.1016/j.bbr.2023.114544

Abstract

Behavioral studies are commonly used as a standard procedure to evaluate anxiety and depression in animal models. Recently, different methods have been developed to improve data collection and analysis of the behavioral tests. Currently available methods, including manual analysis and commercially available products, are either time-consuming or costly. The objective of this study was to improve the collection and analysis of behavioral test data in animal models by developing an image processing program. Eleven behavioral parameters were evaluated by three different methods, including (i) manual detection, (ii) commercially available TopScan software (CleverSys Inc, USA), and (iii) In-housed-developed Advanced Move Tracker (AMT) software. Results obtained from different methods were compared to validate the accuracy and efficiency of AMT. Results showed that AMT software provides highly accurate and reliable data analysis compared to other methods. Less than 5% tolerance was reported between results obtained from AMT compared to TopScan. In addition, the analysis processing time was remarkably reduced (68.3%) by using AMT compared to manual detection. Overall, the findings confirmed that AMT is an efficient program for automated data analysis, significantly enhancing research outcomes through accurate analysis of behavioral test data in animal models.

Keywords: Animal behavior; Anxiety; Automated analysis; Depression; Image processing; Software.

MeSH terms

Animals
Behavior Rating Scale*
Data Collection
Image Processing, Computer-Assisted / methods
Models, Animal
Software*