| Popis: |
Artificial intelligence has the potential to revolutionize health care, yet clinical trials in neurological diseases continue to rely on subjective, semiquantitative and motivation-dependent endpoints for drug development. To overcome this limitation, we collected digital readout of whole-body movement behaviour of Duchenne muscular dystrophy patients (n=21) and age-matched controls (n=17). Movement behaviour was assessed while the participant engaged in everyday activities using a 17-sensor body suit during 3 clinical visits over the course of 12 months. We first defined novel movement behavioural fingerprints capable of distinguishing DMD from controls. Then, we used machine learning algorithms that combined the behavioural fingerprints to make cross-sectional and longitudinal disease course predictions, which out-performed predictions derived from currently used clinical assessments. Finally, using Bayesian Optimization, we constructed a behavioural biomarker, termed the KineDMD ethomic biomarker, that is derived from daily-life behavioural data and whose value progresses with age in an S-shaped sigmoid curve form. By combining an approach that embraces daily life movement motor behaviour with machine learning, our biomarker provides a potential pathway for determining when a new therapy effect occurs or weans off. |
