Digital health metric selection framework

external page This link contains the MATLAB source code for the paper

A data-driven framework for selecting and validating digital health metrics: use-case in neurological sensorimotor impairments. Christoph M. Kanzler, Mike D. Rinderknecht, Anne Schwarz, Ilse Lamers, Cynthia Gagnon, Jeremia Held, Peter Feys, Andreas R. Luft, Roger Gassert, Olivier Lambercy. npj Digital Medicine, 3, 80 (2020).

This code will simulate data for multiple digital health metrics for a reference (e.g., neurologically intact) and a target (e.g., neurologocially impaired) population. Based on these data, the proposed multi-step metric selection framework will be applied and all evaluation criteria and plots from the paper will be generated.

This approach is expected to be applicable for any 1D, discrete digital health metric for that data for a reference (e.g., neurologically intact) and target (e.g,. neurologically impaired) population is available. Also, one of these populations should have test-retest data available as well.
The framework allows the selection of robust metrics that have highest potential for repeatedly assessing impairments based on discriminant validity, test-retest reliability, measurement error, learning effects, and inter-metric correlations.ore metrics, paving the way for their integration into neurorehabilitation trials.
 

external page Download Publication from Nature