MyHeart

Intelligent Biomedical Clothes for Monitoring, Diagnosing and Treatment of Cardiovascular Diseases.
 

Objectives

Multisensor data fusion for realtime feedback in neurological rehabilitation.

Labmedinfo led the development of MyHeart's Concept NR:Neurological Rehabilitation -one of the four applications selected to become a Product Concept. Concept NR provides a telemedicine platform for the early recovery of neurologically-ill patients.

Concept NR platform supports physicians and therapists in the delivery of personalized recovery exercise plans. The recovery process supports both motor rehabilitation exercises, which are monitored with kinestetic strain-sensor garments,and cognitive rehabilitation, based on speech-therapy exercises.

Description of the work

The key challenges for lowering the mortality rate in CVD and their related costs are by successful guiding, informing and motivating the citizen to adapt to a permanently healthier life style and the early diagnosis of acute events.

maglia-sensori.jpg Intelligent clothes, paired with user feedback devices form an integrated MyHeart NeuroRehab system. The project combines novel wearable technologies (textile and electronic sensors, personalised algorithms, on-body computing), real-time feedback, and gesture recognition via sensor-fusion algorithms, in order to make a prototype application for continued care after neurological events.

Available student theses

  • Data mining on biometric signals: multi-sensor fusion.
  • Use of wireless motion sensors for clinical assessment.
  • Implementation of feedback questionnaires on a palm-top device.

Partners


Technology

  • Multi-sensor based biofeedback
    • Wearable strain-gauge sensors
    • MEMS attitude sensors for posture detection
    • Stereophotogrammetry-based calibration
  • Logopedic exercises for cognitive rehabilitation
  • Touch-Screen based UI for feedback and progress visualization
  • Continuous Monitoring, Personalised Diagnosis and Therapy
  • Remote Access and Professional Interaction


Demonstrations

Computer-assisted motor rehabilitation with wearable sensors
Demonstration of the patient workstation supporting motor rehabilitation with a sensorized garment. After an initial training phase, the workstation's algorithms detect subject's movement, verify its performance and provide a feedback in real time. (Flash video with audio, available in English and Italian language. Requires Flash player). 

Avatar animation by MEMS attitude sensor
Attitude sensors allow realtime 3D animation of an avatar. Sensor fusion algorithms will merge data from MEMS sensors in order to complement garment readings.  (Flash video. Requires Flash player).

 

Released Software

Dataset: upper-limb rehabilitation exercises acquired through 29 elastomer strain sensors placed on fabric. Release 1.1 (RAR archived, 10 Mb).

Dynamic time warping algorithms in R: a comprehensive DTW package for the R project. Available on CRAN and R-forge.

 

Further Information (external links)

To the Official MyHeart Site

At a Glance: IS Project Profile

Press Release: Philips Research

 

Related research

Wearable strain sensors for recognizing trunk movements