Mathematical Processing of Electrophysiological Signals for Reproduction by a Hardware-Software Simulator

Mathematical Processing of Electrophysiological Signals for Reproduction by a Hardware-Software Simulator

Margarita  Sidorova
Department of “Biomedical engineering"
Penza state technological university
Penza, Russian Federation
sidorova_mailbox@mail.ru

Nataliya Serzhantova
Department of “Biomedical engineering"
Penza state technological university
Penza, Russian Federation
itmmbspgta@yandex.ru

Anton Syomin
Department of “Biomedical engineering"
Penza state technological university
Penza, Russian Federation
anton.semin.88@inbox.ru

Abstract—The article is devoted to mathematical processing features of electrophysiological signals. Such processing is needed for creating test sequences models as "reference" structures for subsequent reproduction by specialized devices – functional generators and simulators. The article discusses the main features of bio-electric signals, which are the basis for creating test sequences which include: the absence of standards of "norms" for various types of bio-electric signals; heterogeneity and functional complexity of the research object; a large spread of the measured biological object parameters (genetic, sex, age, and other differences); a significant level of noise and some other features. Signal data taken from a bio-object should be presented in a form that takes into account the simplicity, convenience of evaluation, analysis and subsequent interpretation not only for specialists, but also for students and postgraduates engaged in engineering research in the field of biomedicine. The article also presents the justification of choosing the most information-intensive electrophysiological signals for subsequent analysis and mathematical processing, and defines their main measured parameters. An example of an electrocardio signal fragment processed by standard mathematical functions is given as a way to create models of test sequences. The authors of the article offer an original solution to the processing problem – the separation of the electrocardio signal into sections without the separation of individual waves. The test sequences models  of bio-signals taken from a person by this method can be used to evaluate the performance of medical devices and to train students in medical and technical fields.

Keywords—processing, signal, simulator, test sequence, biological, device

1. Introduction

Biological signals (BS) taken from a person are called electrophysiological signals (ES). The result analysis of the ES parameter measurements is allows to choose and apply the best techniques and devices of influencing the bio-object.

The most diagnostically significant signals are: electrocardiogram (ECG), electroencephalogram (EEG) and electromyogram (EMG) ones. The means of recording and reproducing these signals must be tested in a timely manner, and their performance must be evaluated. This assessment method required by high-quality operation of these facilities and complexes. Poor-quality devices can create significant distortions in the acquisition, measuring and analysis of signal parameters and subsequent processing[1, 2].

The reliability of the obtained results, the correctness of the diagnosis and treatment of diseases depend on the quality of functioning and reliability of devices, systems and hardware used in medicine. Generators (both standard and specialized) are usually used as a means of evaluating performance, as well as ES simulators [2-5].

These devices have long been used as means of testing medical facilities and complexes, but they also have a few drawbacks. Generators due to its functional features and hardware structure has a small set of test signals, and simulators are not standard devices. The way out of this situation, according to the authors of the article, is the development and using of simulators capable of reproducing test sequences (TS) of the ESTS not only from standard signal bases, such as, for example, MIT-BIH, but also the original databases of mathematically processed ESTS  models. An example of bio-signal parameters processing and creating such a database is discussed in the article, and the need to systematize the models of ESTS for subsequent reproduction by a hardware-software simulator is also indicated.

1. The goal and the tasks of the study

Currently, despite the variety of reproducing means of ES, in particular, simulators of bioelectric signals, there is no there is no database of TS models of various types of ES, and not just the electrocardio signal. There is also no universal tool that can reproduce the created models when checking the performance and testing of medical devices, as well as for use in the teaching process of medical-technical areas.

The goal of the article is to describe the stages of mathematical processing of the ES form parameters for creating models of the ESTS and the means of their reproduction in the hardware-software simulator form.

• the specific characteristics of the BS are highlighted;
• the most information-intensive ES attributes are determined and a fragment of mathematical signal processing is presented on the example of ECG;
• an example of forming the ESTS model for system simulation of electrophysiological signals, which is based on hardware-software simulator is given;
• recommendations were given for the creation of a database of ESTS and a hardware-software simulator of electrophysiological signals capable of reproducing models.

1. Description of the research subject

The subject of the study is biological signals, more precisely, the method of forming the ES models. A biological signal is kind of signal that is transmitted by transformed currents of biological origin [6].

It should be noted that BS can represent a change in time of electrical parameters or be non-electrical[6, 7].

 A. Relevance of mathematical processing of ES

ES are demonstrate the bio-systems properties, thus interpretation of their parameters and characteristics is useful diagnostic information that can helps explain and recognize various deviations from the norm. The process of deciphering can be implemented by manual effort – visual checkup of the signal in hard copy or on a computer monitor. At once, the complexity of the signal is high and therefore therefore, biomedical signal pre-processing is a prerequisite for extracting diagnostically reliable and adequate information that the signal carries [6, 7].