Proyectos Activos


3D Kinematics for Remote Patient Monitoring

Referencia: ATTRACT RPMD3D

Duración: 20/05/2019 - 19/05/2020

Tipo de proyecto:  Público

Ambito de proyecto: Internacional

Entidades participantes: 

  •   UAB - Computer Vision Center
  •   ULPGC - IDeTIC
  •   University of Applied Sciences and Arts Western Switzerland
  •   Guttmann Institute
  •   Polytechnique Montréal

Lider del proyecto:

  • Cenrte de Visió per Computador (Universidad Autónoma de Barcelona)

Investigador Principal:

  • Alicia Fornés

Investigador Principal ULPGC:

  • Miguel Ángel Ferrer Ballester

Resto de Investigadores: 

  •   Cristina Carmona Duarte
  •   Moises Díaz Cabrera (IDeTIC)


  •   Horizon 2020 ATTRACT Projects


A brain stroke occurs when the flow of oxygen-rich blood to a portion of the brain is obstructed or when a blood vessel leaks or bursts. It is the second leading cause of death in Europe, and 60% of the survivors have different degrees of disability, with a socio-economic impact of the first magnitude for the patient, their environment, the health system and society in general. Thus, it is crucial to find personalized and suitable treatments during stroke rehabilitation.

The main goal of this project is to explore the feasibility of remote monitoring in brain stroke rehabilitation based on the analysis of 3D movements captured with smartbands (a worldwide and non-intrusive technology). We aim to derive from the 3D kinematics an objective estimator of the improvement of the patients’ motor abilities during rehabilitation.

We base our analysis on the Kinematic Theory of Rapid Human Movement, which provides a mathematical description of the movements made by individuals, reflecting the behaviour of their neuromuscular system. It has demonstrated a great potential for monitoring neuromuscular diseases, among others, in terms of the alteration of the ideal parameters. However, it requires robust algorithms to estimate the model parameters with an excellent precision for a meaningful neuromuscular analysis. So far, such algorithms have focused mainly on 1D and 2D movements in a controlled scenario, e.g. pen movements on a tablet computer. This constraint makes the approach unrealistic as a worldwide tool for treating certain cerebrovascular accidents, such as brain strokes.

The objectives are:

1) To adapt the parameter extraction algorithms to 3D movements.

2) To extract and analyse the model parameters obtained from inertial sensors of smartbands.

3) To identify and extract the relevant gestures in continuous 3D movements for their posterior analysis.

4) To implement a proof of concept to validate our research in a real case scenario for brain stroke rehabilitation at the Guttmann Institute.

This explorative work will have a great impact in homecare telehealth tasks. The integration of an analytic tool in a consumer and affordable technology such as smartbands could be used for continuous remote patient monitoring in the rehabilitation stages, improving the medical efficiency and reducing the healthcare costs.

From a technological point of view, the ground breaking novelty of this proposal is the adaptation of the parameter extraction for the kinematic model to a sequence of continuous 3D movements in an unconstrained scenario, paving the way for applications in biomedicine, biometrics, robotics, simulations, video games, human-machine interfaces, etc.

More information: