Multi-purpose Electrostimulator Software

Nuno Costa

, Tiago Araujo

1,2

, Neuza Nunes

and Hugo Gamboa

1,2

CEFITEC, Departamento de Física, FCT, Universidade Nova de Lisboa, Lisbon, Portugal

PLUX - Wireless Biosignals S.A., Lisbon, Portugal

Keywords:

Human Computer Interaction, Electrostimulation, Multi-purpose Software.

Abstract:

Nowadays, most of the software for electrostimulation is made with speciﬁc purposes, and in some cases

they have complicated user interfaces and large, bulky designs that deter usability and acceptability. A novel

Human Computer Interaction framework was developed enabling the end user to conﬁgure and control an

electrostimulator, surpassing the speciﬁc use of several electrostimulator software. In the conﬁguration the

user is able to compile different types of electrical impulses (modes) in a temporal session, and this session

can be actuated in the control. To help the user in creating any type of protocol (session) we devised three

standard impulse generator (rectangular, sin and triangular) and a new way of creating electrical impulses by

drawing, then ﬁtting this data and process it with a mathematical algorithm for ﬁnding simple equations to

describe the data. With it, the user has the possibility to choose the best equation that ﬁts the draw and store

it mathematically structured, thus adding not only a draw editor, but also an equation editor. Therefore, we

provide a tool for clinical, sports and investigation where the user is free to produce their own protocols by

sequentially compile electrical impulses.

1 INTRODUCTION

In 1791, scientist Luigi Galvani showed that electric-

ity, when applied to a frog’s leg, could cause muscle

twitches (Piccolino, 1998). In the intervening cen-

turies, scientists learned much about how Electrical

Stimulation or Electrostimulation (ES) affects mus-

cle tissue and have tried to apply that knowledge to

muscles paralysed by neuromuscular disease to cre-

ate both therapeutic and functional effect. This ﬁeld

Device

Mode

Actuation

Control

Session

Configuration

(Electrostimulator)

(Impulse generator)

(Mode compiler)

Specialized

Professionals

Patients

Basic users

Figure 1: Generic diagram of the HCI.

of medicine is known variously as Electrical Mus-

cle Stimulation or Electromyostimulation (EMS),

Neuromuscular Electrical Stimulation (NMES) and

Functional Electrical Stimulation (FES) electromyos-

timulation, consisting in nerve manipulation through

electrical pulses aiming muscular contraction or sen-

sory response for distinct applications (Malmivuo and

Plonsey, 1996). Nowadays, the applications can be

divided in two main areas:

• Electrotherapy: Rehabilitation (Von Lewinski

et al., 2009); Spinal cord injury, stroke, sensory

deﬁcits, and neurological disorders (with Neural

prostheses) (Cogan, 2008); and many others.

• Physical Conditioning: ﬁtness; active recovery;

optimizing physical performance by improvement

of maximum strength of a muscle (muscular

tonus) in less time (Siff, 1990).

Although ES may hold much promise there are

many technical challenges that need to be surpassed.

Commercial software solutions for electrostimulators

grow every day, but these are often limited by a va-

riety of factors including cost, source code inaccessi-

bility, hardware compatibility, and more (Keller et al.,

2002). Consequently, a strong tradition in scien-

tiﬁc research is to write custom software routines.

While superb for the speciﬁc tasks at hand, these

225

Costa N., Araujo T., Nunes N. and Gamboa H..

Multi-purpose Electrostimulator Software.

DOI: 10.5220/0004241002250230

In Proceedings of the International Conference on Biomedical Electronics and Devices (BIODEVICES-2013), pages 225-230

ISBN: 978-989-8565-34-1

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

Figure 2: Device Conﬁguration for the Electrostimulator used to test the software.

custom solutions rarely offer the ﬂexibility and ex-

tensibility needed for them to be transferable across

platforms, hardware conﬁgurations, and experimental

paradigms without signiﬁcant modiﬁcations. There-

fore, in present time, software/hardware solutions to

provide a device with a multi-purpose platform (sport,

therapy or investigation), a dynamic software, which

enables the user to create their own protocols, is

needed (Suter et al., 2010). To contribute in solving

this necessity, we created a WEB based Human Com-

puter Interaction (HCI) that allows the user to employ

the software in different types of ES applications, pur-

suing a non-speciﬁc software.

This paper further depicts this novel HCI based on

a dynamic Knowledge Acquisition System developed

previously enabling the fast re-edition of conﬁgura-

tions if needed (Costa et al., 2012). The speciﬁc focus

of this tool is to compile electrical impulses in differ-

ent sessions for actuation of an electrostimulator, en-

abling the user to design particular types of protocols.

To surpass this challenge, innovative ways of compile

and create electrical impulses were implemented.

The next sections are divided in: Human Com-

puter Interaction, a resume of the structure developed

in a previous study; Conﬁguration, a description of

the methods provided to the user to compile electrical

impulses; Control, a brief description of the simplic-

ity in actuate a session.

2 HUMAN COMPUTER

INTERACTION

A novel Human Computer Interaction was developed

with the intention of controlling the actuation of elec-

trical devices, like Biosignals Acquisition Systems,

LEDs, motors, between others. Nevertheless, to show

the potential of the tool, the biggest and the ﬁrst chal-

lenge that led us to idealize it, was to control an elec-

trostimulator for different types of applications. To

demonstrate it, this article shows the capabilities of

the software in controlling an electrostimulator and

the contribution for Electrostimulation.

The HCI is integrated in a WEB based software

for biosignals acquisition and processing from PLUX

- Wireless Biosignals, S.A., enabling the control of a

generic device. In this way, actuation, acquisition and

processing will be possible in a closed-loop cycle.

In Biomedicine, the conﬁguration of a device

should only be performed by clinical experts, where

they can adapt sessions to each patient. Overly com-

plicated user interfaces and large, bulky designs can

deter users from operate the device on a day to day

basis. For example, home health care devices should

have sessions previously programmed enabling the

patient to only start or stop a session prepared for

him, and restrict the software to this purpose because

patients can’t be allowed to conﬁgure or change any

BIODEVICES2013-InternationalConferenceonBiomedicalElectronicsandDevices

226

Figure 3: Example of the default impulses with the restrictions speciﬁed in the Device Conﬁguration: a)square, b)sine and

c)triangle. The arrows mark the direction in which the interactive points can be moved. In each row are presented examples

of the respective impulse after moving the interactive points.

parameter in the software. Thus, to economize time,

simplify interfaces, automate the device actuation and

separate this tool for experts and non-experts the soft-

ware was divided in Conﬁguration and Control. Ba-

sically, as shown in Figure 1, to control we need to

conﬁgure the electrostimulator, and create temporal

sessions, i.e. sequence of impulse modes, for each

channel of that device. Consequently, the conﬁgura-

tion is hierarchically structured in device, mode and

session. In the control we only need to choose the de-

vice and connect to it, if connected chose the session

to setup the hardware, then start the session. The stop

can be manual or automatic (end of the session).

3 CONFIGURATION

3.1 Device Requisites

In the case of an electrostimulator, some speciﬁc op-

tions need to be satisﬁed. The speciﬁcations for the

potentials and limits of ES is dependent of the ap-

plication and the electrostimulator. The requisites of

the electrostimulator we used for this project, are:

MAC Address, "00:07:80:4b:fb:be"; Pulse amplitude,

0-100 mA (1 mA step); Pulse Width, 0-500 µs (5 µs

step); Pulse frequency, 1-200 Hz (1 Hz step); Number

of channels, 2; Number of modes, 4; Waveform type,

Rectangular, triangular, sinusoidal, customized wave-

form (constant potential with no offset). In this way,

Figure 2 represents the Device Conﬁguration with the

above requirements.

3.2 Mode

After the speciﬁcation of the device requisites the user

can project a new mode for a speciﬁc device. The

mode permits four types of impulse conﬁguration:

square, rectangular pulse that can be changed within

the data limits; sine, sin pulse that can be changed

within the data limits; triangle, triangular pulse that

can be changed within the data limits; draw, a novel

and innovative impulse generator. The user is also

able to choose the frequency (impulse per second) by

moving a slider.

In the three default types square, sine and triangle,

the user have to click and pull the interactive points.

Examples of the three types are shown in Figure 3,

each row represents an example of the respective im-

pulse after moving the interactive points. These types

allow the end user to create a lot of different impulses,

but to be able of creating any type we had to devise a

new form of impulse generation, the draw method.

In this new type, the user has the possibility to

Multi-purposeElectrostimulatorSoftware

227

Figure 4: Conﬁguration of the mode "On" for the Electrostimulator. Initial version of the software.

draw an impulse within the data limits, from this draw

we retrieve 100 points that will be processed by an al-

gorithm. The idea behind this algorithm it is to search

within the 100 points that seem connected to each

other, then propose a series of simple equations to de-

scribe the links. The best are selected, tweaked, and

again tested against the data. Next, the algorithm re-

peats the cycle over and over, until it ﬁnds equations

that have a good probability of modulate the data.

Then, by choosing one of the equations, the user can

see a graph with 100 points based in the equation and

compare it to the drawing. If it ﬁts the user objectives,

save the equation, if not, the user can tweak the equa-

tion manually, and see the effects, or just save the 100

points from the draw. This algorithm will have large

beneﬁts when ﬁnished, because it provides a mecha-

nism of storing the draw data mathematically struc-

tured, in this way, the user will have the possibility

of re-editing the data by changing some parameters in

the stored equations. So, instead of just providing a

draw method to create impulses, an equation method

is also available, with great relevance by simplifying

the impulse generation when the impulse can be de-

scribed by an equation and only few parameters need

to be changed.

This algorithm is still in the initial stage and is

based in the work of Schmidt and Lipson, the Eu-

reqa, a highly praised symbolic regression program

(Stoutemyer, 2012).

In Figure 4 is possible to view an example of how

a mode can be conﬁgured within the data limits for x

axis 0 - 500 µs and for y axis -100 - 100 mA . The ﬁg-

ure shows the creation of "On" mode for the electros-

timulator, a rectangular pulse with 500 µs, amplitude

100 mA, and 20 Hz of pulse frequency, i.e. 20 pulses

per second.

3.3 Session

The user can create a new session for a speciﬁc

device. A session is a sequence of modes, and

each channel of a device can have one session pro-

grammed. A device has a maximum number of pro-

grammable modes, but they can be repeated inﬁnitely.

This mechanism is very important to program stim-

ulation protocols, automating the control. Figure 5

shows an example of session conﬁguration. Primar-

ily, the deﬁnition of the modes that will be used, four

is the maximum that the hardware can memorize. Af-

terwards, the creation of a session for the channels re-

quired, in this case, only one channel is programmed

with two modes: On, 300 seconds activated and, Off,

during 100 seconds (Off is a pulse with 500 µs and

amplitude 0 mA). This sequence is repeated during

half an hour by activating the loop for that speciﬁc

session.

Additionally, we remark that each conﬁguration

presented is not static, input new ﬁelds, re-edit or

delete is fast and simple due to the dynamic Knowl-

edge Acquisition System implemented for the HCI.

A very important mechanism to optimize and adapt

quickly the software, especially the conﬁguration

BIODEVICES2013-InternationalConferenceonBiomedicalElectronicsandDevices

228

Figure 5: Conﬁguration of one channel session with the modes "On" and "Off". Initial version of the software.

(Costa et al., 2012).

4 CONTROL

Control is the division for actuation and acquisition of

biosignals. In here, the user ﬁrst needs to choose the

device, in our case the electrostimulator, and the ses-

sion to program. When the device is ready, start/stop

of the session is feasible. In the course of the session

is possible to acquire real time biosignals and syn-

chronize them with the session.

Figure 6 presents the Control environment, di-

vided in Setup and Actuation. In the Setup the user

picks one of the devices conﬁgured, connects using

the MAC address deﬁned in the Device Conﬁgura-

tion, then chooses the session to program in the Hard-

ware. If the hardware is correctly set up, start/stop

and information of the session for each channel are

available within the Actuation.

The control area will have many device sessions

templates. Therefore, a quick approach by the user

will enable the simple selection of a pre-deﬁned ses-

sion and start the actuation immediately.

5 CONCLUSIONS

A strong tradition in neurophysiology research is to

write custom software routines. These software are

speciﬁc and rarely offer the ﬂexibility and exten-

sibility needed for them to be transferable across

platforms, hardware conﬁgurations, and experimen-

tal paradigms without signiﬁcant modiﬁcations. Also,

most of the times these software have confuse, large

and bulky interfaces that can deter users from using

the software. Electrostimulation is one of the areas

that suffers from these problems. For these reasons,

and pursuing usability and acceptability, we are devel-

oping a HCI that has a great potential to be transfer-

able across different types of electrical device control,

and to demonstrate it, we are able to produce different

types of sessions to control the actuation of an elec-

trostimulator.

For this purpose we provide mechanisms to con-

ﬁgure and control a device. The most important

mechanisms are: the separation in Conﬁguration and

Control; Conﬁguration, where clinical professionals

can conﬁgure a device with sessions for each channel

trough the compilation of electric impulses; Control,

where the user just needs to choose the device and

session (set up the device) then simply start/stop the

session prepared in the conﬁguration; the sequence of

conﬁguration (Device, Mode, Sessions), that supplies

the user with a structured conﬁguration; the tools in

Mode, enabling the end user to shape the impulses at

is own desire, and with the draw, plus equation algo-

rithm, implemented the user will be able to produce

any kind of impulse and store information mathemat-

ically structured; the compilation of modes in Ses-

sion allowing the creation of temporal sessions for

Multi-purposeElectrostimulatorSoftware

229

Figure 6: Control environment using the Electrostimulator deﬁned in the conﬁguration and session "protocol1".

the actuation of an electrostimulator, i.e. permits to

program stimulation protocols. Thus, creating a dy-

namic, ﬂexible way of automate the actuation of stim-

uli.

Depending on the stakeholder the combination

Conﬁguration-Control will change. Conﬁguration

will be enabled, disabled or will have different levels

of restrictions. In the Control area different devices

and sessions will be programmed by default for a fast

approach by the user (setup the device then start/stop

the session). Ideally, the various envisioned users

should be consulted on whether they would be open

to use such a software package and, if so, under what

restrictions. Interviews or questionnaires will be suit-

able instruments for this. We hope this could be the

foundation for building a business model and further

develop ideas on how to penetrate the market.

We also provide a tool to generate dynamic con-

ﬁgurations, that are fast and simple to re-edited.

Therefore, if for example the conﬁgurations described

in this article are not optimized to conﬁgure a speciﬁc

electrostimulator we can easily optimized them.

With this powerful tool we provide a user-friendly

software solution with a multi-purpose platform for

sport, therapy and research.

REFERENCES

Cogan, S. (2008). Neural stimulation and recording elec-

trodes. Annu. Rev. Biomed. Eng., 10:275–309.

Costa, N., Araújo, T., Nunes, N., and Gamboa, H. (2012).

Knowledge acquisition system based on json schema

- implementation of a hci for actuation of biosignals

acquisition systems. In Cabello, E., Virvou, M., Obai-

dat, M. S., Caldeirinha, R. F. S., Ji, H., and Vergados,

D. D., editors, SIGMAP, pages 255–262. SciTePress.

Keller, T., Popovic, M., Pappas, I., and Müller, P. (2002).

Transcutaneous functional electrical stimulator “com-

pex motion”. Artiﬁcial organs, 26(3):219–223.

Malmivuo, J. and Plonsey, R. (1996). Bioelectromagnetism,

volume 34. Peter Peregrinus LTD.

Piccolino, M. (1998). Animal electricity and the birth of

electrophysiology: the legacy of luigi galvani. Brain

research bulletin, 46(5):381–407.

Siff, M. (1990). Applications of electrostimulation in phys-

ical conditioning: a review. The Journal of Strength &

Conditioning Research, 4(1):20.

Stoutemyer, D. (2012). Can the eureqa symbolic regres-

sion program, computer algebra and numerical analy-

sis help each other? Arxiv preprint arXiv:1203.1023.

Suter, B. A., O’Connor, T., Iyer, V., Petreanu, L. T., Hooks,

B. M., Kiritani, T., Svoboda, K., and Shepherd, G.

M. G. (2010). Ephus: multipurpose data acquisition

software for neuroscience experiments. Frontiers in

Neural Circuits, 4:100.

Von Lewinski, F., Hofer, S., Kaus, J., Merboldt, K.,

Rothkegel, H., Schweizer, R., Liebetanz, D., Frahm,

J., and Paulus, W. (2009). Efﬁcacy of emg-triggered

electrical arm stimulation in chronic hemiparetic

stroke patients. Restorative Neurology and Neuro-

science, 27(3):189–197.

BIODEVICES2013-InternationalConferenceonBiomedicalElectronicsandDevices

230