Scholarlink

Abstract <p>The behavioral test “Morris Water Maze” is a useful device to study spatial learning, behavioral reactions and memory of small laboratory animals. Within the framework of the joint project between MLIT and LRB JINR aimed at creating an information system, we have designed a web service to automate the analysis of experimental data related to the behavioral test “Morris Water Maze.” In data analysis, the automation tasks are concerned about the analysis of video data. Meanwhile, the development of convenient tools can significantly reduce the research time and the human factor influence. Here we present the results of the development of a web service which is designed to annotate and classify the data for the trajectories of rodent movements in the “Morris Water Maze” behavioral test. The functionality of the service enables a person to monitor the correctness of the constructed trajectory, classify the trajectory based on a deep learning approach and obtain a number of characteristic parameters of the rodent movements during the experiment. The web service was developed and deployed based on the ML/DL/HPC ecosystem of the HybriLIT Heterogeneous Computing Platform.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Algorithm for the Analysis of the Laboratory Animal Trajectories in the “Morris Water Maze” and Its Implementation as a Web Service

  • 期刊论文

  • 发表日期：

    2025年10月25日

查看全文
  • T. Zh. Bezhanyan,
  • S. Shadmehri,
  • O. I. Streltsova,
  • M. I. Zuev,
  • M. Yu. Bondarev,
  • I. A. Kolesnikova,
  • Yu. S. Severiukhin,
  • D. M. Utina

摘要

Abstract

The behavioral test “Morris Water Maze” is a useful device to study spatial learning, behavioral reactions and memory of small laboratory animals. Within the framework of the joint project between MLIT and LRB JINR aimed at creating an information system, we have designed a web service to automate the analysis of experimental data related to the behavioral test “Morris Water Maze.” In data analysis, the automation tasks are concerned about the analysis of video data. Meanwhile, the development of convenient tools can significantly reduce the research time and the human factor influence. Here we present the results of the development of a web service which is designed to annotate and classify the data for the trajectories of rodent movements in the “Morris Water Maze” behavioral test. The functionality of the service enables a person to monitor the correctness of the constructed trajectory, classify the trajectory based on a deep learning approach and obtain a number of characteristic parameters of the rodent movements during the experiment. The web service was developed and deployed based on the ML/DL/HPC ecosystem of the HybriLIT Heterogeneous Computing Platform.