HoloTeach - a virtual therapy guidance system for personalized rehabilitation after stroke

The project aims to develop an advanced digital rehabilitation system for post-stroke trunk control, integrating motor learning principles, adaptive difficulty, and innovative VR technologies. The system combines VR exercises with a sensor-equipped mobile seat to capture real-time stability metrics.

Key data

Description

Background

After a stroke, regaining trunk control is essential for recovery but remains insufficiently supported by digital tools.

Objectives

Develop an innovative digital rehabilitation system to improve trunk control post-stroke. The system combines VR-based exercises with a sensor-equipped mobile seat (“T-Chair”) capturing real-time stability data such as sway area, direction change, jerk, acceleration, velocity, and range of motion. These metrics feed adaptive algorithms that automatically adjust exercise difficulty to individual performance.

Approach

Instrumented Seat (ZHAW-IMES): Integration and analysis of T-Chair sensor data to enable dynamic therapy adaptation; iterative testing with patients and therapists ensures practical usability.
Adaptive Difficulty Algorithms (UZH): Development of real-time algorithms combining VR and sensor data. Speech and icon interfaces allow autonomous patient interaction.
Motor Learning & Clinical Translation: Neuroscientists and clinicians (UZH, Valens, Cereneo) apply motor learning principles to design protocols for subacute and chronic rehabilitation phases.
Usability & Co-Design: Participatory development involving patients and therapists through iterative feedback cycles to enhance intuitiveness, accessibility, and motivation.
Movement Analysis (ZHAW-IPT, UZH): Motion capture and EMG studies with healthy subjects and stroke patients to evaluate and refine adaptive algorithms based on movement patterns.