Automatic detection of foot-strike onsets in a rhythmic forelimb movement.

Rhythmic movement is the fundamental motion dynamics characterized by repetitive patterns. Precisely defining onsets in rhythmic movement is essential for a comprehensive analysis of motor functions. Our study introduces an automated method for detecting rat's forelimb foot-strike onsets using deep learning tools. This method demonstrates high accuracy of onset detection by combining two techniques using joint coordinates and behavioral confidence scale. The analysis extends to neural oscillatory responses in the rat's somatosensory cortex, validating the effectiveness of our combined approach. Our technique streamlines experimentation, demanding only a camera and GPU-accelerated computer. This approach is applicable across various contexts and promotes our understanding of brain functions during rhythmic movements. • Defining onsets in rhythmic motion is crucial for an analysis of motor functions. • We introduced automated detection of foot-strike onsets using deep learning tools. • This detection method combines two deep learning techniques. • This method enables accurate detection of onsets across many animal subjects. • This method promotes understanding of brain functions during rhythmic movements. [ABSTRACT FROM AUTHOR]