A Quantitative Kinematic Movement Biomarker Characterizing Neurodevelopmental Disorders*

When we make reach-directed movements to an object, like a cup, there is an infinite number of possible trajectories that an arm could take to go from its initial position to the final target. These are not given by solving Newton’s equations but are controlled by the brain. Humans show highly stereotyped random trajectories with hand velocity profiles that appear smooth to the naked eye. However, by using high definition motion capture MEM sensors we have found that at finer millisecond time scales, away from naked eye detection, the kinematic variables have clear random peak fluctuations that were unknown previously. We have been able to directly connect, in an individualized way, these random fluctuations to the nature of the Neurodevelopmental Disorders (NDD) of children with autism or attention hyperactive disorders. These random peak fluctuations are very different, though, in healthy individuals. In this lecture I will describe the nature of experiments we carry out, the importance of fully filtering out the external electronic noise from neuronal noise. Me measure, angular velocity, linear acceleration, and the time derivative of the acceleration, known as jerk of importance in human kinematics. Based on the data measured over thousands of arm movement trials we developed a detailed statistical analysis that allowed us to uncover quantitative NDD biomarkers that a posteriori agreed with the clinical diagnoses for each individual studied. We further hypothesize that the brain tries to minimize the associated stochastic kinematic variable fluctuations describing the motions. The identified biomarkers quantitatively determine the illness severity, that could be used as a diagnostic tool by clinicians.

Torres, E. B., José, J. V. el al. Frontiers in Integrative Neuroscience, 7, 2013.
Wu, D., José, J. V., Nurnberger, J. I., and Torres, E. B. Scientific Reports (Nature), 8(1):614, 2018.