Reducing robotic upper-limb assessment time while maintaining precision: a time series foundation model approach.

Abstract

Purpose: Visually Guided Reaching (VGR) on the Kinarm robot yields sensitive kinematic biomarkers but requires 40-64 reaches, imposing time and fatigue burdens. We evaluate whether time series foundation models can replace unrecorded trials from an early subset of reaches while preserving agreement with full-session estimates of standard Kinarm parameters.

Methods: We analyzed VGR speed signals from 461 stroke and 599 control participants across 4- and 8-target reaching protocols. We withheld all but the first 8 or 16 reaching trials and used ARIMA, MOMENT, and Chronos models, fine-tuned on 70% of participants, to forecast synthetic trials. We recomputed four kinematic features of reaching (reaction time, movement time, posture speed, max speed) on combined recorded plus forecasted trials and compared to full-length references using ICC(2,1).

Results: Chronos forecasts increased ICC values for all parameters ([Formula: see text]) when combining only 8 recorded trials with forecasted trials, achieving agreement comparable to that obtained using 24-28 recorded reaches ([Formula: see text]). MOMENT yielded intermediate gains, while ARIMA improvements were minimal. Across cohorts and protocols, synthetic trials replaced reaches without significantly compromising feature reliability.

Conclusion: Foundation-model forecasting can greatly shorten Kinarm VGR assessment time. For the most impaired stroke survivors, sessions drop from 4-5 min to about 1 min while maintaining agreement with full-session Kinarm parameter estimates. This forecast-augmented paradigm promises efficient robotic evaluations for assessing motor impairments following stroke.