Fit4MedRob: Design and Preliminary Validation of a Modular Robotic Framework for Hemispatial Inattention Rehabilitation
Microprocessors and Microsystems · 2026-11 · 105357
DOI: 10.1016/j.micpro.2026.105357 ↗Kurz gefasst
Fachartikel in der Zeitschrift Microprocessors and Microsystems, 2026-11. Verfasst von Bruna Maria Vittoria Guerra und 5 weiteren Autor:innen. Die Arbeit nennt 20 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Ministero dell'Istruzione dell'Università e della Ricerca.
Abstract (Originalsprache)
The Fit4MedRob initiative (“Fit for Medical Robotics”) is a large-scale project funded by the Italian Ministry of University and Research under the Italian Complementary National Plan, involving 25 partners across academia, clinical centers, and industry. Within this framework, Mission 2 – Activity 5 (Match-Making #53) focuses on the development of robotic solutions for the rehabilitation of post-stroke hemispatial inattention (HI), a condition characterized by impaired awareness of one side of space. This work presents the design and preliminary validation of a modular robotic framework based on the TIAGo platform, a collaborative mobile manipulator developed by PAL Robotics. The proposed architecture integrates six functional modules: Navigation, Patient Identification, Patient Interaction, Patient Monitoring, Exercise Delivery, and Evaluation. The system is designed to support structured rehabilitation sessions while enabling the acquisition of quantitative sensor-based measurements. Preliminary validation was conducted with healthy participants under controlled conditions and focused on selected technological components of the framework, including autonomous navigation, patient identification, speech-based interaction, markerless head-orientation estimation, and upper-limb motion tracking. Ground-truth kinematic measurements were obtained using a stereophotogrammetric system where applicable. Navigation enabled the robot to establish an interaction configuration compatible with subsequent identification, although systematic differences were observed between onboard and reference estimates of subject distance and alignment. Patient identification was successfully achieved within the intended interaction range, while the speech-based interaction module reached an overall intent-classification accuracy of 92.6%. Markerless head-orientation estimation showed high agreement with the reference system for yaw, whereas larger errors were observed for pitch. Upper-limb tracking captured the main spatial and temporal characteristics of robot-guided movements, although angular alignment showed greater variability.
Förderung
- Ministero dell'Istruzione dell'Università e della Ricerca
Zitieren
APA
Guerra, B. M. V., Soldi, R., Sozzi, S., Russo, L., Schmid, M., & Ramat, S. (2026). Fit4MedRob: Design and Preliminary Validation of a Modular Robotic Framework for Hemispatial Inattention Rehabilitation. Microprocessors and Microsystems, 105357. https://doi.org/10.1016/j.micpro.2026.105357
Vancouver
Guerra BMV, Soldi R, Sozzi S, Russo L, Schmid M, Ramat S. Fit4MedRob: Design and Preliminary Validation of a Modular Robotic Framework for Hemispatial Inattention Rehabilitation. Microprocessors and Microsystems. 2026:105357. doi:10.1016/j.micpro.2026.105357
BibTeX
@article{Guerra2026,
author = {Guerra, Bruna Maria Vittoria and Soldi, Roberto and Sozzi, Stefania and Russo, Leo and Schmid, Micaela and Ramat, Stefano},
title = {Fit4MedRob: Design and Preliminary Validation of a Modular Robotic Framework for Hemispatial Inattention Rehabilitation},
journal = {Microprocessors and Microsystems},
year = {2026},
pages = {105357},
doi = {10.1016/j.micpro.2026.105357}
}
EUVIMED