Setting digital psychiatry in motion: towards dynamic digital markers for digital phenotyping

TL;DR

The paper introduces a dynamic approach to digital phenotyping that captures how mental health patterns change over time, offering a better understanding of psychopathology.

Contribution

The novel contribution is the proposal of dynamic digital markers that model time-varying aspects of mental disorders.

Findings

Traditional entropy-based measures fail to capture temporal dependencies in digital phenotyping.

Dynamic digital markers better reflect the evolving nature of mental disorder processes.

The dynamic approach improves the representation of variability in regulatory mechanisms of psychopathology.

Abstract

Digital phenotyping uses data from smartphones and wearables to extract behavioural and biosocial markers of psychopathology in situ. Traditional entropy-based measures capture static system properties that neglect temporal dependencies critical to psychiatric phenomena. We propose a “dynamic” approach to the modelling of digital data capturing the time-varying aspects of processes of mental disorders. We defend that the resulting dynamic digital markers better capture variability in regulatory mechanisms of psychopathology.

Digital phenotyping uses information from smartphones and wearable devices to track patterns in behavior and physiology related to mental health. Most current methods summarize this data in ways that miss how experiences change over time. We propose a new, dynamic approach that focuses on how patterns evolve moment by moment. By capturing these changes, our method aims to better reflect how mental disorders involve shifting and unstable processes of regulation in everyday life.

Figures