Smartphone Typing Dynamics for Assessing Hand Function in Psoriatic Arthritis: A Proof-of-Concept Study

Psoriatic Arthritis (PsA) is a chronic inflammatory disease affecting both the peripheral and axial skeleton, significantly impacting patients' quality of life. Diagnosing and monitoring PsA remain challenging, as they often rely on subjective assessments by experienced healthcare professionals. In this study, we introduce a novel method for assessing hand function impairment, with emphasis on fine motor skills, in individuals with PsA by developing digital biomarkers (dBMs) that analyze patterns formed during natural typing interactions with touchscreen smartphones. The assessment is conducted using a smartphone-based typing test enabled by a custom keyboard for passive capturing of keystroke parameters and typing metadata and outputs statistics of keystroke dynamics variables, such as key hold time and flight time between key taps. Typing data were acquired from a cohort of 16 clinically verified PsA patients, among whom 11 exhibited low disease activity (LDA) and five exhibited high disease activity (HDA), and nine healthy controls (HC). Typing-based dBMs combined with logistic regression, achieved an 85% classification performance for distinguishing HC from LDA (AUC[95%CI] : 0.85[0.84, 0.85]), indicating that typing dynamics could be exploited for early PsA symptom screening. Overall, the experimental results revealed statistically significant measures and promising discriminative capabilities, underscoring the potential of integrating dBMs into clinical practice for enhanced PsA assessment.Clinical relevance- This proof-of-concept study highlights the potential of digital biomarkers, derived from natural typing interactions on touchscreen smartphones, to enable objective, accessible, and early detection of PsA-related hand function impairment.