AI-Driven Framework for Enhanced and Automated Behavioral Analysis in Morris Water Maze Studies

Area of Science:

• Neuroscience
• Behavioral Science
• Artificial Intelligence

Background:

• The Morris Water Maze (MWM) is crucial for assessing spatial learning and memory, especially in neurodegenerative disease research.
• Traditional MWM analysis methods have limitations in capturing complex animal behaviors.
• There is a need for more precise and reliable MWM data evaluation.

Purpose of the Study:

• To develop and validate a novel AI-based automated framework for processing and analyzing Morris Water Maze (MWM) test videos.
• To enhance the accuracy and reliability of behavioral analysis in MWM experiments using machine learning.
• To provide a robust tool for studying neurodegenerative disorders like Alzheimer's disease.

Main Methods:

• Developed an AI pipeline including video preprocessing, animal detection via Convolutional Neural Networks (CNNs), and trajectory tracking.
• Implemented a novel concentric circle segmentation approach alongside traditional quadrant-based division of the MWM pool.
• Extracted 32 behavioral metrics per zone and utilized machine learning classifiers (e.g., random forest, neural networks) with feature selection for classification tasks.

Main Results:

• The AI framework successfully processed MWM videos, extracting detailed behavioral features.
• Classification tasks differentiating younger and older animals showed significant performance improvements.
• Integration of features from concentric zone analyses notably enhanced classification accuracy.

Conclusions:

• The AI-based automated framework offers a robust, precise, and reliable solution for MWM data processing.
• This approach significantly improves the analysis of spatial learning and memory in behavioral neuroscience.
• The enhanced MWM analysis is critical for advancing research into neurodegenerative disorders.