Machine Learning-Informed Digital Measures Improve the Sensitivity and Translational Relevance of In Vivo Neurodegenerative Disease Models

Neurodegenerative diseases pose significant global health challenges, affecting millions of individuals worldwide. Amyotrophic Lateral Sclerosis (ALS) is a progressive disorder impacting motor neurons in the spinal cord and brain, leading to muscle weakness, and loss of motor control. The average life expectancy following a diagnosis is only 2-5 years. These statistics underscore the urgency and importance of studying ALS to understand the mechanisms of neurodegeneration and develop effective therapeutic strategies. Many of the current in vivo readouts used to assess neurodegeneration are subjective, time-consuming, and labor intensive. Rapid advances in sensor technologies and computational capabilities provide a unique opportunity to enhance the value of animal studies. Complementing standard measures with continuous measures of behavior and physiology would provide a more dynamic, biologically-, and clinically relevant characterization of disease progression and therapeutic effects.

Here we used the B6.Cg SOD1-G93A model of ALS and the B6.Cg-Pvalbtm1(cre)Arbr Fxnem2Lutzy Fxnem2.1Lutzy/J model of Friedreich’s Ataxia to assess machine-learning informed digital measures compared to standard measures of disease progression including body weight, rotarod, neuroscore, compound muscle action potential (CMAP), and neurofilament light chain (NfL) in home cages outfitted with computer vision cameras with infrared detection capabilities allowing continuous monitoring of socially housed individual animals throughout both light and dark cycles over 28 weeks. In both models, disease phenotype was identified earlier using digital measures compared to most standard endpoints. Additional insights were observed when evaluating sleep behavior.

This innovative approach, demonstrated through objective and quantitative longitudinal assessment of digital measures of animal behavior, could provide valuable insights into neurodegenerative disease and contribute to the development of effective interventions. Our work has the potential to significantly advance our understanding of neurodegenerative disease and pave the way for novel therapeutic strategies, thereby addressing a critical need in global health.