#	Pagina
attuale pagina	/open-h2020/projects/209075/results.html
-1	/open-h2020/projects/199722/results.html
-2	/open-h2020/projects/199025/results.html
-3	/open-h2020/projects/221840/index.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - MyoBan (wireless Body Area Networks for high density MYOelectric neurorehabilitation technologies)

Teaser

Summary

The goal of the project MYOBAN is the development of a new system for the recording and wireless transmission of multiple surface HD EMG (High Density Electromyography) electrodes in real-time. The system will overcome the current limitations based on wired technology for the translation of the HD EMG systems into clinical applications.
The increased number of elderly people in developed countries will require focusing on innovative solutions to overcome the consequences of motor disability due to cardiovascular or degenerative diseases. This means that motor impairment as, decrease in muscle strength, low manual dexterity and difficulty in the coordination of multiple muscle groups will affect a bigger number of people in the next future. Therapies can increase muscle tone, improve coordination of multiple muscles and re-activate the brain areas involved in planning and execution of voluntary movements. Rehabilitation treatments are essential to provide independence of self-care activities, walking with or without assistive devices and improve cognitive skills.
Nevertheless, classic therapies do not heal motor impairment with consequent limitations in independent living. The limited understanding of human control of movements and the impossibility to evaluate reliably the motor deficit at the muscle and neural level are the main reasons for the poor outcome of classic therapies. In the last few years, a new generation of assisting devices for rehabilitation has been developed to overcome the limitations of conventional rehabilitation procedures. The main idea behind these new systems is the detection and interpretation of the muscle activation signals (EMG) and provide biofeedback to the physical therapist for tuning the rehabilitation protocols in real-time.
The project focused on the development of a new generation of wireless surface HD EMG systems that will be used in future neurorehabilitation technologies. The devices include innovative processing of the HD EMG recordings, to reduce the amount of data transmitted, and a communication standard which respect the requirements for body area network. The final aim was the production of small HD EMG system that can be placed on the surface of multiple muscles of the patients with minimal invasiveness.

Work performed

In the first part of the project, a review of the state of the art for wireless surface EMG systems has been conducted. The review has been directed checking the available standards for wireless communication that are compatible with the transmission of biosignals. In literature, several standards have been exploited for the transmission of biomedical signals. Among them, two possible candidates have been identified for the specific application of the project: the IEEE 802.11 and the IEEE 802.15.6. Both standards can reach data rate compatible with the application and can be exploited for wearable devices. Additionally, a search was performed among the patented devices with similar characteristics to the proposed project. It has been found that the most important companies that are producing clinical and research devices are increasing their interest in wireless surface EMG systems. Nevertheless no high-density wireless surface EMG system (>16 channels) implementing a standard communication protocol was found available on the market.
The most optimized communication standard for the proposed project was found to be the IEEE 802.15.6, that is specific for wireless body area networks. This standard can include a large variety of transmission bandwidths for different applications. The IEEE 802.15.6 has the advantage that may be used with much lower power transmission (10mW) compared to other standards.
Unfortunately, the standard is relatively new and, there are no commercially available hardware implementations that may be easily included in the proposed system. Therefore, its use requires the implementation of the entire standard in a prototype board.
For this reason, the project focused mostly on the improvement of an already existing system developed by the OT Bioelettronica in the last year. The system adopted is based on the well-established standard IEEE 802.11.
The second part of the project focused on the optimization of the features included in the standard IEEE 802.11 and on a simplified framework for the compression of the sampled surface EMG signals.
This implementation selected provides the flexibility to select the main transmission bandwidths (2.4 or 5 GHz), channels, data rate, Access Point/Station mode, modulation type, power management, security protocol, etc. The performance of the system was analyzed, in terms of interference, when multiple devices were active at the same time. The tests were done transmitting simulated data or previously collected surface EMG signals by OTB. Using the default parameters of the hardware implementation, the level of interference was quite high already with only two devices active at the same time. This resulted in a considerable reduction of data rate during the transmission and/or the packet loss of collected samples. For these reasons, several months were used optimizing the combination of the features in the wireless module in order to overcome some of these interference problems. This approach provided the possibility to reliably transmit sampled surface EMG signals from two wireless modules active at the same time, but it could not overcome the interference level when more modules were active. It was clear the necessity to include source coding of the sampled data in order to reduce the overall data rate and therefore decreasing the interference level among the wireless modules. Consequently, a low-complexity compression algorithm has been designed. This approach permited the use of four devices concurrently. This was an important outcome of the project since it provided the possibility to implement a network of four wireless modules that could sample high-density surface EMG data from four muscle groups at the same time in human subjects.
In the last part of the project, it was implemented a set of Matlab numerical simulations to understand the efficiency of a second generation of the wireless system using a simplified version of the IEEE 802.15.6 standard. These

Final results

The project results showed the possibility to implement a network of multilple wireless modules that could sample high-density surface EMG data (64 channels) from different muscle groups at the same time in human subjects. Furthermore, in the near future, a new generation of surface HD EMG systems can be implemented using a communication standard specific for on-body applications.