Glossary
Digital biomarker
An objective, measurable feature derived from digital data — sensors, devices, software — that reflects a biological or clinical state.
A biomarker is any measurable indicator of a biological state: blood pressure, a lab value, a score on a scale. A digital biomarker is one computed from data collected through digital means — wearable sensors, a phone's microphone, a gait mat, software that logs how a task is performed. Examples in neurology include stride-length variability from a gait sensor, speech pacing from recorded voice, or reaction-time distributions from a cognitive test on a tablet.
Digital biomarkers are attractive because they can be captured frequently, cheaply and outside the clinic, which makes them well suited to tracking slow change. They are only useful once validated: a candidate feature must be shown to be reliable, to correlate with a clinically meaningful outcome, and to change in a way that matters. Most of the work is in the capture and the validation, not in the model.
Why it matters to our work
Our platform captures gait, balance, cognition, speech and medication-response signals longitudinally so that candidate biomarkers can be mined and then validated with clinicians — the foundation of our neuroanalytics work.