|
Abstract:
|
Here we present two collaborative efforts utilizing supervised learning to develop tools for preclinical and early clinical biomarker discovery, which provide robust quantification of translatable biomarkers derived from the sleep architecture. First, we describe a robust automated method for polysomnography utilizing EEG & EMG signals. It works across species, mice, rats and non-human primates, and across experiments and labs. Second, we present an approach which combines measurements of activity and heart rate variability to improve sleep-wake activity endpoints and enable scalable longitudinal monitoring in clinical trials.
|
