350 rub
Journal Biomedical Radioelectronics №3 for 2020 г.
Article in number:
Remote diagnostics of upper respiratory tract diseases based on the analysis of speech signal parameters
DOI: 10.18127/j15604136-202003-09
UDC: 612.789.1
Authors:

S.N. Kirillov – Dr.Sc. (Eng.), Professor, Head of Department «Radio Control and Communication»,  Ryazan State Radio Engineering University n.a. acad. V.F. Utkin; Academician of the International Academy  of Communications; Corresponding Member of the Academy of Military Sciences, Honored Worker of the Russian Higher School;

E-mail: kirillov.lab@mail.ru

D.Yu. Mamushev – Candidate, Department «Radio Control and Communication», 

Ryazan State Radio Engineering University n.a. acad. V.F. Utkin

E-mail: mamushev_du@bk.ru

Abstract:

Statement of the problem. A method of remote diagnostic decision-making decisions in diseases of the throat and nose of the upper respiratory tract (URT proposed based on the analysis of the parameters of the speech signal (SS) at the output of the channels for transmitting speech information (CTSPI).

Aim of the work – to investigate the impact of CTSPI on the results of diagnosis of of the upper respiratory tract (URT) diseases by MS parameters, to clarify the models of the method of group accounting of arguments (mgua) for the diagnosis of relevant diseases, as well as to clarify the algorithm for classifying these diseases based on mgua.

Results. The study of diagnostic efficiency of the obtained models and GMDH were carried out on the basis of an assessment of the probability of false rejection of a correct diagnosis of the patient (type I errors) and false diagnosis (type II errors) in accordance with the methodology described in, without the use of KPI and the passage RS through KPI. The analysis of the data obtained during the study shows that the use of remote diagnostics via cellular phone channels allows us to solve the problem of determining the preliminary diagnosis of a patient with ENT disease. But at the same time, the probabilistic characteristics of the resulting solution deteriorate accordingly.

The possibility of remote diagnostics by telephone channels of cellular communication of patients with ENT diseases is shown. The possibility of remote diagnosis of diseases of the upper respiratory tract (URTI), such as the curvature of the nasal septum, polyposis sinusitis, rhinitis, tonsillitis, based on the analysis of the parameters of the speech signal (RS) at the output of the speech information transmission channels (CPI), is evaluated. The influence of speech codecs such as AMR-NB, AMR-WB (G.722.2), CELP (G.728), G.723.1, ADPCM (G.726), which are widely used in voice transmission channels of the main cellular communication operators, is studied. such as MTS, Beeline, Megafon, Tele2, etc. It is established that when passing RS through the KPI for 2G networks, the probabilities of correct diagnosis are reduced by 7-12%, and in the case of 3G, 4G standards – by 48%. At the same time, the probabilities of false diagnosis increase for networks of the 2G standard by 5–12%, and in the case of 3G, 4G standards by 0.5–2%. 

Practical significance. The results confirm the practical use of the proposed method for making a preliminary diagnostic decision about the patient’s disease.

Pages: 68-74
References
  1. Kirillov S.N., Luk'yanov D.I. Ocenka vozmozhnosti vyyavleniya psihoemocional'noj sostavlyayushchej rechevogo signala na vyhode kanalov peredachi rechevoj informacii. «Rechevye tekhnologii». 2016. № 1–2. S. 13–19 (In Russian).
  2. Rechevye kodeki. [Elektronnyj resurs]: Retrieved from http://celnet.ru/vocod.php 04.04.20 (In Russian).
  3. Kirillov S.N., Kuchurkin A.N., Mamushev D.Yu. Metod podderzhki prinyatiya diagnosticheskih reshenij pri zabolevaniyah verhnih dyhatel'nyh putej po parametram rechevogo signala. Biomedicinskaya radioelektronika. 2019. T. 22. № 4. S. 68–76 (In Russian).
  4. Kirillov S.N., Shustikov O.E, Mamushev D.Yu. Identifikacionnaya znachimost' staticheskih i dinamicheskih priznakov rechevogo signala v sistemah raspoznavaniya diktorov. Vestnik RGRTA. 2000. Vyp.7. S. 92–97 (In Russian).
  5. Kirillov S.N., Mamushev D.Yu. Hahulin S.S. Raspoznavanie i identifikaciya rechevyh komand na osnove nelinejnoj modeli golosa diktora. Sb. tr. XVIII sessii Rossijskogo akusticheskogo obshchestva. 2006. S. 55–57 (In Russian).
  6. Ivahnenko A.G. Dolgosrochnoe prognozirovanie i upravlenie slozhnymi sistemami. Kiev: Tekhnika, 1975. 312 s. (In Russian).
Date of receipt: 26 мая 2020 г.