Detecting and Quantifying Ataxia-Related Motor Impairments in Rodents Using Markerless Motion Tracking With Deep Neural Networks

In this study we evaluate the application of video-based markerless motion tracking based on deep neural networks for the analysis of ataxia-specific movement abnormalities in rodent models of cerebellar ataxia. Based on a small amount (<100) of manually labeled video frames, markerless motion tracking enabled the extraction of movement trajectories and parameters characterizing ataxia-specific movement abnormalities. In the first experiment, we analyzed videos of 6 shaker and 4 wildtype rats and were able to reproduce the5 Hz tremor frequency in the shaker rat without the usage of a force plate. In the second experiment, we investigated a spinocerebellar ataxia type 3 (SCA3) mouse model (6 mice aged 3 months and 3 mice aged 9 months) in a beam-balancing task. By establishing a parameter for the assessment of rhythmicity of gait (RoG), we not only found a significantly higher RoG in wildtype mice compared to affected SCA3 mice aged 9 months, but were also able to reveal a significantly lower than typical RoG in SCA3 mice aged 3 months which exhibit no abnormalities in visual inspection. These prototypical results suggest the capability of the presented methods for the application in upcoming therapeutic intervention trials to identify subtle changes in movement behavior.