Grant manager:

Lecturer Dan Marius Dobrea, M.Sc., Ph.D.

Working project:

• A Non-Contact Bio-Instrumental System for: Quantitative Evaluation/Analysis of Parkinson Patients and Body Language Interpretation

This research is supported with the help of:

• Romanian National University Research Council, Contract no: 33371/29.06.2004, theme 18 (2004), theme 19 (2005), code NURC 493, 2004-2005
• TIara grant - Texas Instruments Article Reward Assignment - that supply the tools:
  • TMS320C6416 DSK
  • LVDS evaluation board, SN65LVD387EVM

Short Descriptions of the Project

The goal of this project is to build a bioinstrumental machine to be used in:

• quantitative evaluation of the head movements in Parkinsonian patients;
• intelligent human computer interface, where the system is able to understand the human body language.

The project will:

• improve the performances, overcome the drawbacks and enlarge the field of applicability of the already built prototype system [Dobrea, 2002];
• obtain new relevant outcomes for the actual state of the research in the reference field;
• design the bioinstrumental system as a self-contained unit that will be easily integrated in a more complex and intelligent system designed for the identification of the biopsychic human states.

The team will capitalize the experience acquired in the previous implementations of the prototype system [Dobrea, 2002] and will totally re-design it, in order to obtain a new device with extended field applicability and better performances. For this, the research will have as objectives:

• to re-design the mechanical part of the equipment - so as to assure an increased accuracy in the determination of the body position in the 3D input space;
• to implement the control system of the mechanical part, to create the image acquisition and preprocessing units and the analysis subroutines - all built inside a DSP modul, connected to a PC that will receive only the analysis results (an improvement beside the prototype, where the PC had to deal with all these jobs);
• to write the subroutines to extract the following information: head position, its trajectory description, gestures and postures of the subjects and their classification.

The technical solutions are addressing to the following key problems:

• electro-mechanical interface and control;
• digital image signal processing for object extraction and description;
• feature extraction for postures classification and head trajectory analysis;
• communication protocols;
• subject pattern recognition of contextual postures used to describe the user's state.
  

[Dobrea, 2002] Dan-Marius Dobrea, A New Type of Sensor to Monitor the Body Torso Movements Without Physical Contact, EMBEC'2002, Proceedings of Second European Medical and Biological Engineering Conference, December 4-8, 2002, Vienna, Austria, IFMBE Proceedings, Vol. 3, Part 1, pp. 810-811, ISBN 3-901351-62-0