A software-hardware solution for analysis of phono-cardiograms Текст научной статьи по специальности «Компьютерные и информационные науки»

A SOFTWARE-HARDWARE SOLUTION FOR ANALYSIS OF PHONO-CARDIOGRAMS

DOI 10.24411/2072-8735-2018-10309

Sergey V. Bachevskiy,

St.Petersburg State University of Telecommunications St. Petersburg, Russia, rector@sut.ru

Andrey G. Vladyko,

St.Petersburg State University of Telecommunications St. Petersburg, Russia, vladyko@bk.ru

Vladimir S. Uss,

St.Petersburg State University of Telecommunications St. Petersburg, Russia, uss_wl@mail.ru

Gleb G. Rogozinsky,

St.Petersburg State University of Telecommunications St. Petersburg, Russia, gleb.rogozinsky@gmail.com

Keywords: telemedicine, phonocardiogram, measurement stand, electronic phonendoscope, LabVIEW, Hilbert transform, multi domain model.

At the present time the active development of the digital telemedicine technologies, being the part of common Industrial 4.0 paradigm, plays key role for the improvement of quality of medical care, especially in the regions of the Russian Arctic zone, where the digital equipment of remote diagnostics of patients is widely used. Modern smart sensors can carry out the primary diagnosis measurements without medical specialist's supervision. For example, smart phonen-doscopes with learning features can recognize various heart diseases by the results of analysis of measured phonocardiograms (PCG), thus considerably reduce medical assistance in the sense of physical presence of trained personnel, and in some cases can save a patient from the lethal outcome. Meanwhile, this demands for development of specialized software-hardware solutions for patient's statistical data gathering and corresponding modification of diagnostics algorithms, able to process such data and adjust the algorithms of diagnosis classification for the purposes of accuracy increasing. Because of the described research, the stand for measurement and analysis of PCG had been designed. The proposed solution is a working prototype of specialized diagnostic complex, enabling integration with software-hardware solutions based on LabVIW, as an industrial standard within various measurement-related applications including telemedicine. As the conceptual basis for the development of given methods we used the multi domain communication model, which allows to describe the aspects of the physic, informational and cognitive domains in the single terms. In particular, this model allows formalization of various transitions and transformations in-between different media, which is significant in the development of cyber-physical measurement and control systems.

Information about authors:

Sergey V. Bachevskiy, Dr. Sc. Tech. Prof. of Communications Theory and Radiotechnics Department, St.Petersburg State University of Telecommunications, St. Petersburg, Russia

Andrey G. Vladyko, Ph.D., Head of Science-Research Institute of Communication Technologies, St.Petersburg State University of Telecommunications, St. Petersburg, Russia

Vladimir S. Uss, Ph.D. Associated prof. of Radioelectronics Design and Production Department, St.Petersburg State University of Telecommunications, St. Petersburg, Russia

Gleb G. Rogozinsky, Ph.D., Head of Medialabs, St.Petersburg State University of Telecommunications, St. Petersburg, Russia Для цитирования:

Бачевский С.В., Владыко А.Г., Усс В.С., Рогозинский Г.Г. Программно-аппаратное решение для исследования фонокардиограмм // T-Comm: Телекоммуникации и транспорт. 2019. Том 13. №9. С. 50-57.

For citation:

Bachevskiy S.V., Vladyko A.G., Uss V.S., Rogozinsky G.G. (2019). A software-hardware solution for analysis of phono-cardiograms. T-Comm, vol. 13, no.9, pр. 50-57.

Introduction

Nowadays, electronic phonendoscQpes are increasingly used for analyzing the state of patients [1 j. The possibility of obtaining phonograms of the heart (phonocardiograms) and recording them on electronic media demands developing an appropriate diagnostic complex. Such complex can be used for people working in long-period expeditions or living in hard-to-reach areas where highly qualified medical professionals are typically absent and where there is an acute problem of providing timely and skilled care to patients, including heart attacks prevention and treatment. However, the development and debugging of disease recognition algorithms lead to the appropriate stand design, allowing collecting statistical data of phonocardiograms and improving the accuracy of recognition,

1. The Multi Domain Model of Communications

First, we generalize the Rogozinsky-Sotnikov model of communications [1], mainly proposed for the soniilcation systems, to the data gathering/exchange purposes described in the paper. According to the model, any kind of activity can be understood as an exchange between three main domains - Physical (PD), Informational (115) and Cognitive (CD). The I'D is typically concerned with the energy processes and the interaction of material world objects. The situation analysis and intellectual activity are ihe products of mental and psychic activity of the CD. The ID is the area for the circulation of data used in the CD, representing the objects, phenomena and process of the PD.

At the domain borders, the corresponding interfaces perform the information interaction between different elements of the system. The finite number of states, represented by its own thesaurus, can characterize each object/subject of the system.

Thus, the object A of a PD with corresponding thesaurus slates, mapped onto the multiplicity of informational representations {/l}-'1 of the thesaurus

(1)

In other words, the information is transferred, when the signal transmitting an image (notion) from the varifold thesaurus of the source system A into the varifold thesaurus of the target system B is changed. The information is received when a new image of the source is formed within the varifold thesaurus of the target system.

(2)

where - mapping operators between different domains,

i.e. PD. ID or CD

thesauri of a target signal and a source signal.

In detail, we can write (2) as following

(<,f f

(3)

Thus the mapping between the corresponding domains is actually the operation of information impact between entities of the

domains, expressed in the discovery of maximum conformity between elements of thesauri ■

Information impact is the influence of the "source" A on the state of the "target" R, which manifests itself in the change of the image ( B), seen in the variety of elements within ihe thesaurus

of the "source" (Since the "source" and "target" Cn thesauri are different, the original image in the internal thesaurus

of (Band the image of { B pA in the "source" thesaurus are also different. This can lead to errors, or inaccurate representation of the object in the thesaurus of the user.

Information exchange is the receiving and transmission of

signals leading to the mutual alteration of images (A and £ . { BpA of the exchange participants. This can be caused by alteration (expansion) of the participants' thesauri £A and £,b-

Information interaction is the mutual change of images of

ow n systems of (A)^A and { B)%B, leading to the change of images {A )-'f and { B)bA in the other participants.

Information system (IS) is a system containing "information" and providing it to the user. A necessary condition is as follows: "The necessary components of an IS are: the user and the potential information". A sufficient condition is "The user and the potential information form an IS". IS arc made up of elements which are information images (A ) of the real (material and immaterial) entities A and possess information significance.

Information significance is the property of representing the entity, which requires a descriptive method containing a set of basic "meanings", immanent to the entity. "Information significance" should not be confused with "value" or "usefulness"; these properties are related to the users and the possibility of satisfying their need of information.

The formalized set of "meanings" is the & thesaurus. The entity item has a number of discernible slates, which are perceived by the observer as a set of images of an object, each having its own "meaning". The number of states determines the potential information carried by the object. When the observer acquires an image of the object (by means of perception and recognition), the potential information is actualized on the basis of the information representation of the object. The potential presence of information in system A is determined by the set of discernible stales of the system and the varifold system thesaurus E^.

Perception of the information transferred occurs when the receiver R acquires a new image of system A in the varifold receiver thesaurus

User V is a person, object or process capable of perceiving

images {S)^s and possessing its own receiver thesaurus

2. The Object-to-Listener Chain Links

The thesaurus design for any sound gathering/measurement system depends on lots of factors. That is to say, it is not only an issue of the pure sound design research, but aiso a subject of multiple limitations, which are to be estimated and assumed to prevent the loss of the meaningful information, or its undesired misrepresentation.

The main unit of any soniilcation system is a human operator, or listener. Thus, the properly designed soniilcation system

should be conformed to the operator's hearing charaeteristics. The latter are well known from the research of Z wicker, Beranek and other acousticians and physiologists. After them, we typically outline ail ability to sense an intensity of the signal, which correlates with the volume of the sound; an ability to sense the frequency, which correlates with the pitch and ability for spatial localization of the signal's source due to our binaural hearing. We are also able to analyze the complex structure of the signal, which correlates with the timbre.

Assuming the auditory display, we conditionally outline five source-target links within the object (physical or virtual) and the listener, i.e. object - generator, generator - display, display -medium, medium - listener's sensory system (LSS), LSS - listener's cognition. Each link transforms the previous thesaurus of the information carried with the sound.

If we assume an existence of some object A as a thing-in-itself with its own thesaurus, which includes all possible states of

i \cA ■

a given object, we can define it as (A) , i.e. object A, represented in its own thesaurus .

Every transfer of the object's representation from one media to another, or from one domain to another, e.g. from physical domain to the informational, implies the object's representation in another thesauri, i.e.

(4)

The Multi-Domain Model of communications (MDM), first defined by Sotnikov [2J, provides the universal solution for describing the various processes happen in the (different domains, i.e. Physical (PD), Informational (ID), or Cognitive (CD). The model provides a flexible and abstract approach to describe the transformation of the information inside and on the edge of the domains. Using such approach, we can describe any transformation of signal or object at any step.

First, an object-to-generator path Is divided into two cases. If the sonificaied object is a part of the PD, the process of mapping its thesaurus onto the sound generator described by (2),

{Af^{{Aff >{(.4)'}". (5)

So firstly, an object A of the PD should be transformed to corresponding informational representation, or the entity of the

/ a \ p

ID ) , given in the informational thesaurus . Next

goes the changing of the thesaurus within the ID. i.e. the initial informational representation of the physical object is being adapted to the purposes of sonification.

In the second ease, where the object is a virtual-only entity and exists in a cyber-world, the (2) becomes (3)

(Cf —f (C)' . (6)

The second link, or the generator - display link, describes the signal degradation at passing the path to the speaker or auditory display. Typically, the modern digital equipment will not induce any significant distortion, though in the case of long lines or

noise induction the initial message can be a subject of misrepresentation. This operation lakes part completely ¡11 the ID.

The third link represents the technical acoustical limitations of a speaker system. For instance, if the developed auditory display constructed as a halo-like body connected to the headrest of an operator's chair, with a group of small speakers inside, we should not rely much on the low frequencies due to the limited frequency response of such speakers, so the thesaurus objects should be carefully designed and filtered. The same can be addressed to some wearable units with the small-sized speakers. From the sound design position, such procedure can be understood as a special-purpose sound mastering tasks. From the MDM point, it is yet another intradomain operation.

The fourth link, i.e. medium - LSS, represents the aspects of architectural acoustics, which include the reverberation issues, the absorption of sound, the speaker and listener positions, etc. At this stage, the sounds, which are played through speakers, can lose their legibility and sharpness, the high frequency bands can be sufficiently attenuated because of surfaces and people, the lows can be misheard due to the standing waves or undesired position of the listener. Also sufficiently strong localization issues can be identified because of incorrect speaker or listener placement. At this stage, we should also consider ergonomical limitations, caused by the acoustic peculiarities of operator's working space, i.e. the ambient noise level etc. If the working place is polluted by some noise, it can mask subtle sounds.

The psychoacoustic limitations include several limitations caused by our human auditory system, including different types of masking. Besides masking, the partial hearing loss in adult ear should also be considered. Such limitations demand careful design of a sound set, especially in the ease of simultaneous play of several sounds.

All latter links, i.e. generator - display, display - medium, medium - listener's sensory system (LSS), correspond to the processes or transformations, which take part in the ID. We can group those links under some higher level abstraction, but we would like here to name all possible principal links on the path from object to listener's cognition, since some limitations, which we arc going to describe later, are localized in the given links.

So, we can formally define the described operations through the changing of the initial thesaurus oflD entity.

{Cf. (7)

The final LSS-to-listener's brain link brings the signal to the CD, where the user finally recognizes its meaning, i.e. the initial signal or object after being routed through the links of different domains and transforms, is represented in the user thesaurus of CD .

(cf^ilcff m

3. The hardware components of phonoeardiogram analysis complex

The hardware part of the presented solution includes electronic phonendoscope, a National Instruments myDAQ data acquisition unit, a personal laptop computer with an installed Lab VIEW software (Fife I).

r \

7TT

When analyzing phonoeardiograms, these functions together provide additional information that allows detailing the noise features and transients in the signal of the heart muscle.

The complete block diagram of the program for filtering and processing phonoeardiograms is given on Figure 7. It includes the following blocks:

1) the N1 myDAQ data acquisition device configuration block;

2) the low-pass filtering unit (the 7th order Bessel filters were used);

3) filtering options selection block;

4) a unit for calculating the power spectrum of the signal at the output;

5) the frequency and phase response calculation block;

6) the average output signal calculation block;

7) a unit for obtaining a spectrogram using the two-dimensional Gabor filter;

8) a peak detector unit for measuring the number of heartbeats;

9) the heart pulses calculator;

10) the l lilbert transform unit;

11 ) the amplitude limiter unit;

12) N1 myDAQ device configuration block to output information through the sound card output.

5. The User Interface design

The software interface consists of the main window, divided into four areas (Fig. 8), which display the following:

1 ) the waveform of the original signal obtained as a result of heartbeat measurement procedures {top left);

2) the processed signal and its amplitude envelope (bottom left);

3) the heartbeat signal (top right);

4) spectrogram of the heart muscle activity (bottom right).

All graphs in LabVIEW are interactive. Thus user can change the scales, coloring schemes, styles etc.

Conclusion

As an outcome of the work, a software-hardware solution was created for the research of the phonocardiograms. It allows recording and reproducing the captured data, developing and testing new algorithms for the recognition of various heart diseases. During the further research, we plan to put most of the processing algorithms directly into the memory of the electronic phonendoscope module to increase the operability in patients' diagnosing.

References

1. Sotnikov A.D., Rogozinsky G.G. (2017). The multi domain infocommunication model as the basis of an auditory interfaces development for multimedia informational systems. T-Comm. 2017. Vol. 11. № 5, pp. 77-82.

2. Sotnikov A.D. (2004). Principles of the Applied Area Analysis in Healthcare Infocommunication Systems. Proceedings of Higher Educational Establishments in Communications, no. 171, pp. 174-183.

3. Abbas, K.A., Bassam, R. (2009), Phonocardiography Signal Processing. Morgan and Claypool, pp. 218.

4. Springer D„ Brcnnan, T. (2016). Automated signal quality assessment of mobile phone-recorded heart sound signals: Automated signal quality assessment of mobile phone-recorded heart sound signals. Journal of Medical Engineering & Technology', 40(7-8). pp. 342-355.

5. Baran. E.D. (2009). LabVIEW FPGA. Reconfigured measurement and contrailing systems, Moscow: DMK Press. 448 p. (in Russian)

6. Magda, U.S. (2012). LabVIEW: Practical course for engineers and developers. Moscow: DMK Press. 208 p. (in Russian)

ПРОГРАММНО-АППАРАТНОЕ РЕШЕНИЕ ДЛЯ ИССЛЕДОВАНИЯ ФОНОКАРДИОГРАММ

Бачевский Сергей Викторович, СПбГУТ им. проф. М. А Бонч-Бруевича, Санкт-Петербург, Россия, rector@sut.ru Владыко Андрей Геннадьевич, СПбГУТ им. проф. М. А Бонч-Бруевича, Санкт-Петербург, Россия, vladyko@bk.ru Усс Владимир Станиславович, СПбГУТ им. проф. М. А. Бонч-Бруевича, Санкт-Петербург, Россия, uss_wl@mail.ru Рогозинский Глеб Гендрихович, СПбГУТ им. проф. М. А. Бонч-Бруевича, Санкт-Петербург, Россия, gleb.rogozinsky@gmail.com

Аннотация

В настоящее время активное развитие цифровых телемедицинских технологий, в контексте общей парадигмы Индустрии 4.0, имеет ключевое значение для улучшения качества оказания медицинской помощи, особенно в регионах Арктической зоны РФ, для которой особенно актуальны вопросы удаленного взаимодействия врач-пациент для диагностирования пациентов. Современные умные сенсорные системы телемедицинского назначения способны проводить первичную постановку диагноза без участия врачей-специалистов. В частности, обучаемые электронные фонендоскопы способны распознавать болезни сердца по снятым фонокардиограммам и, тем самым, значительно сократить время оказания помощи больным, а в ряде случаев спасти от летального исхода. Однако, для эффективной реализации таких методов необходимы специализированные стенды, позволяющие обрабатывать статистические данные о пациентах и модифицировать алгоритмы распознавания болезней в интересах повышения точности диагностирования. В результате работы был создан стенд для снятия и анализа фонограмм сердечной мышцы. Представленный в работе стенд является прототипом специализированного диагностического комплекса и способен к интеграции с программно-аппаратными решениями на базе СКМ LabView, являющейся индустриальным стандартом в различных сферах, в том числе и в телемедицине. В качестве концептуальной основы для построения представленных методов был применена мультидоменная модель коммуникаций, позволяющая в единых терминах описать сущности физического, информационного и когнитивного доменов. В частности, данная модель позволяет формализовать описание различных преобразований и переходов между различными средами, что имеет существенное значение при проектировании киберфизических контрольно-измерительных систем.

Ключевые слова: телемедицина, фонокардиограмма, измерительный стенд, электронный фонендоскоп, LabVIEW, преобразование Гильберта, мультидоменная модель.

Литература

1. Sotnikov A.D., Rogozinsky G.G. The multi domain infocommunication model as the basis of an auditory interfaces development for multimedia informational systems. T-Comm 2017. V. 11. Тщю 5. pp. 77-82.

2. Sotnikov A.D. Principles of the Applied Area Analysis in Healthcare Infocommunication Systems. Proceedings of Higher Educational Establishments in Communications, no.171, 2014, pp. 174-183.

3. Phonocardiography Signal Processing. Morgan and Claypool publisher, Abbas K. Abbas and Rasha Bassam, April 2009, pp. 218.

4. Springer D., Brennan T. Automated signal quality assessment of mobile phone-recorded heart sound signals: Automated signal quality assessment of mobile phone-recorded heart sound signals. Journal of Medical Engineering & Technology, 2016.

5. Баран ЕД. LabVIEW FPGA. Реконфигурируемые измерительные и управляющие системы. М.: ДМК Пресс, 2009. 448 с.

6. Магда Ю.С. LabVIEW: Практический курс для инженеров и разработчиков. М.: ДМК Пресс, 2012. 208 с.

Информация об авторах:

Бачевский Сергей Викторович, д.т.н., профессор кафедры теоретических основ связи и радиотехники, СПбГУТ им. проф. М. А. Бонч-Бруевича, Санкт-Петербург, Россия

Владыко Андрей Геннадьевич, к.т.н., директор научно-исследовательского института "Технологии связи", СПбГУТ им. проф. М. А. Бонч-Бруевича, Санкт-Петербург, Россия

Усс Владимир Станиславович, к.т.н., доцент кафедры конструирования и производства радиоэлектронных средств, СПбГУТ им. проф. М. А. Бонч-Бруевича, Санкт-Петербург, Россия

Рогозинский Глеб Гендрихович, к.т.н., доцент кафедры радиосвязи и вещания, СПбГУТ им. проф. М. А. Бонч-Бруевича, Санкт-Петербург, Россия

f I л