Category Archives: Biotechnical and medical systems

MATHEMATICAL MODELS FOR PREDICTING AND EARLY DIAGNOSIS OF DISEASES CAUSED BY ELECTROMAGNETIC FIELDS OF LOW-FREQUENCY RADIO FREQUENCY RANGE


UDC 616.5-002.4
DOI:10.26102/2310-6018/2020.29.2.032

N.A. Korenevsky, A.V. Titova, T.N. Govorukhina, D.A. Mednikov

The paper proposes mathematical models for predicting and diagnosing diseases provoked by exposure to electromagnetic fields of the radio frequency range, which make it possible to control the current state of a person in order to make further decisions about possible correction of body functions, if necessary. Given the incomplete and fuzzy description of the studied class of diseases, soft computing technology was chosen as the basic mathematical apparatus, and, in particular, the synthesis methodology of hybrid fuzzy decision rules, which has proven itself in solving problems with a similar data structure and type of uncertainty. The selected synthesis method allows us to take into account the multiplicative effect of exposure to the human body of electromagnetic fields (EMF) of various modality and intensity, taking into account other endogenous and exogenous risk factors. For powerful and stable EMFs, it is proposed to use a modification of well-known models obtained for industrial power grids. To assess the effect of low-intensity, unstable electromagnetic fields of the radio frequency range on the human body, it is proposed to use fuzzy tabular models and a number of indicators sensitive to the action of the electromagnetic field of the radio frequency range. Such indicators include the state of attention, memory, thinking, as well as the dynamics of changes in the energy state of biologically active points associated with the pathology under study. On the example of electric train drivers, mathematical models for predicting and early diagnosis of the appearance and development of diseases of the nervous system are obtained. It is shown that if additional information about the health status of the subjects is used with electromagnetic risk factors, then confidence in the correct prognosis reaches 0.85, and in the presence of early stages of diseases of the nervous system – 0.95.

Keywords:chronic obliterating diseases of lower limbs, theory of latent variables measurement, optimal treatment regimens.
Full text:
KorenevskySoavtors_2_20_1.pdf

IMPROVING THE QUALITY AND EFFICIENCY OF IDENTIFICATION OF SPECIAL STATES OF MONITORED OBJECTS BASED ON THE DEVELOPMENT OF MATHEMATICAL AND SOFTWARE FOR PROCESSING COMPUTER IMAGES USING LARGE DATABASES

UDC 004.932.2
DOI:10.26102/2310-6018/2020.29.2.030

V.A. Vasilchenko, V.L. Burkovsky

The relevance of the study is due to an increase in human diseases, which are associated with significant socio-economic damage and give a significant burden on health. According to WHO recommendations, a disease prevention system should include prevalence assessment, correction, and risk factor management (WHO, 2009). A special place in this set of measures is occupied by the mass disease monitoring system, both a mechanism for assessing the situation and the need for implementing preventive measures, and a method for monitoring the effectiveness of implemented preventive measures. In this regard, this article considers the creation of an algorithm for processing images of a computer tomography scan of a human lung using software. The leading method to study this problem are neural networks. The article presents a convolutional neural network model of Chexnet X-ray processing developed by scientists from Stanford University. An algorithm for developing a mechanism for analyzing images based on modern x-ray images of organs – computed tomography images, which are obtained using a complex software and hardware complex with ultra-sensitive detectors for recording x-ray radiation, as well as an extensive software package that allows you to obtain images with high spatial resolution, is considered. The developed algorithm is implemented on the basis of the Densenet convolution network, the depth of which is 201 layers. Changes were made to it in the form of using the ReLU activation function (short for English rectified linear unit), which can significantly speed up the learning process and at the same time significantly simplify calculations. As a result, the developed convolutional neural network helps the continuity of data collection, which allows to improve the process of strategic decision-making, to develop action programs in the field of public health.

Keywords:computer image processing, convolutional neural network, ReLU activation function, disease diagnosis.

Full text:
VasilchenkoBurkovsky_2_20_2.pdf

DEVELOPMENT OF A SYSTEM OF MASS MONITORING SPECIAL CONDITIONS

UDC 004.042
DOI:10.26102/2310-6018/2020.29.2.029

V.A. Vasilchenko, V.L. Burkovsky

The relevance of the study is due to the high level of lung disease according to WHO. The annual mortality from chronic lower respiratory diseases is 3 million, and from lung cancer 1.7 million. According to information from the Ministry of Health of the Russian Federation, early diagnosis and planning of preventive measures based on it will significantly reduce the mortality rate from lung diseases and improve the quality of life of the population. To implement the tasks of mass processing of medical information, increase the effectiveness of treatment and prophylactic measures to detect lung diseases in the early stages, it is proposed to create a software package. The developed software makes it possible to assess the situation in the studied territorial area and monitor both the state of health and the effectiveness of measures taken.The article presents methods for automating the analysis of laboratory analysis data, as well as research data of a computer tomograph.The materials of the article are of practical value for medical institutions, allowing you to identify lung pathologies in the early stages, as well as for decision centers, where the quality of medical services is assessed by age and gender groups, both in each region individually and at the state level whole.

Keywords:computer image processing, convolutional neural network, ReLU activation function, disease diagnosis.

Full text:
VasilchenkoBurkovsky_2_20_1.pdf

DECISION SUPPORT SYSTEM FOR DETERMINING THE DOSAGE OF MEDICATIONS IN THE TREATMENT TECHNOLOGY OF PREECLAMPSIA OF PREGNANT WOMEN


UDC 004.891
DOI:10.26102/2310-6018/2020.29.2.017

M.V. Grankov, I.A. Tarasova

The problem of preeclampsia is one of the urgent in modern obstetrics, since this disease is the most common and serious complication of pregnancy, and the problem of treating severe forms of preeclampsia is one of the most difficult in obstetric anesthesiology and resuscitation. The high mortality rate is based on the lack of accurate knowledge about the pathogenesis of the disease, which depends on many factors, diagnostic criteria, which leads to inadequate therapy and various complications, depending on the timeliness and method of delivery, the volume of anesthetic and resuscitation care. Therefore, the study of methods for constructing automated and expert systems using modern methods of artificial intelligence and allowing to increase the effectiveness of the treatment of preeclampsia of pregnant women is relevant. This article discusses the development of a decision support system for determining the dosage of medications in the treatment technology of preeclampsia of pregnant women based on membership functions of several arguments. As a result of experimental tests, it was found that the relative deviation of the dosages calculated by the decision support system from the dosages established in the comparative tests by a qualified doctor does not exceed five percent. At the same time, the use of the results of the work made it possible to increase the number of severe patients served by one resuscitation doctor by at least two times, by reducing the time to establish a diagnosis.

Keywords:decision support system, diagnostics, treatment technology, preeclampsia of pregnant women, membership function of several arguments.

Full text:
GrankovTarasova_2_20_1.pdf

MODIFICATION OF GENETIC ALGORITHM WITH ADAPTIVE CROSSOVER SWITCHING


UDC 681.3
DOI:10.26102/2310-6018/2020.29.2.009

Y.A. Asanov, S.Y. Beletskaya, Al-Saedi Mohanad Ridha Ganim

The aim of this work is to develop a modification of the adaptive genetic algorithm based on switching crossover in accordance with the degree of elitism of individuals in the population. Despite the enormous amount of research done in the field of evolutionary calculus in recent years, algorithms of this class today have a high prospect of modification. The main aim of research is carried out in order to improve the convergence rate of algorithms (to obtain high-performance optimization methods) and increase the accuracy of the solutions obtained. In the article, for the adaptive tuning of the crossover operator, the concepts of discrete and continuous degree of elitism of individuals are used. In addition, an elitism score is used to adjust the probability of a mutation. This modification has a serious advantage superiority in test problems which are traditionally used to analyze the efficiency of genetic algorithms. The test set used was a quadratic function with three variables, a Rosenbrock function, a step function, a complex fourth-order function with noise, and the Sheckel function. The results of comparing classical genetic algorithms with algorithms using the considered crossover and mutation tuning strategies are presented. An analysis of the results of a computational experiment is presented.

Keywords:genetic algorithm, switching crossover, adaptive mutation tuning, elitism, evolutionary calculus.

Full text:
AsanovSoavtors_2_20_1.pdf

APPLICATION OF CORRELATION ANALYSIS TO IDENTIFY FACTORS FROM A WOMAN’S ANAMNESIS INFLUENCING THE RESULTS OF PREGNANCY OBTAINED BY INF


UDC 519.234.3
DOI:10.26102/2310-6018/2020.28.1.027

S.L. Sinotova, O.V. Limanovskaja, A.N. Plaksina, V.A. Makutina

This article discusses the search for a statistical relationship between diseases of the genitourinary system, chronic diseases, surgical interventions and other data on the anamnesis, a woman’s heredity and pregnancy outcome obtained using assisted reproductive technologies (IVF). The study is conducted with the aim of developing a mathematical model for predicting pregnancy and assessing the health of a child conceived using ART (IVF) at the stage of planning. The conclusions are based on data on 338 women and the diagnoses of their 402 children at the stage of the maternity hospital. A research was made for the effect of 56 binary signs on the outcome of pregnancy, described by 38 characteristics. To identify significant factors, a correlation analysis was performed using Fisher’s exact test, Chi-square test, and using interval estimates of the shares, and the Z-criterion for the difference of two shares. As the outcomes, the terms and methods of delivery available to the patient group under consideration, as well as the diagnoses of children at the stage of the maternity hospital were selected. To assess the strength of the relationship, Cramer’s V is applied. The result of the analysis is the identification of 56 significant factors and 35 significant correlations, which will be taken into account in the future for the development of the regression model.

Keywords: correlation analysis, p-value, analysis of small data, assisted reproductive technologies, Cramer’s V, Chi-square test, Fisher’s exact test, mathematical statisticscorrelation analysis, p-value, analysis of small data, assisted reproductive technologies, Cramer’s V, Chi-square test, Fisher’s exact test, mathematical statistics.

Full text:
SinotovaSoavtors_1_20_1.pdf

MODELING OF THE PROCESS OF INTERMOLECULAR INTERACTION FOR THE SELECTION OF ANTIDOTES NEUTRALIZING THE TOXIC IMPACT ON THE COMPONENTS OF THE CELLULAR MEMBRANE


UDC 501, 004.942, 615.91
DOI:10.26102/2310-6018/2020.28.1.028

I.M. Azhmukhamedov, L.I. Zharkikh

To study the effects of toxicants on a living organism and the selection of effective antidotes, studies are usually carried out in vivo or at least in vitro, which is a very laborious and costly process. In addition, such studies are not always possible because of ethical considerations. Experiments on living creatures in the most countries are very strictly regulated by law. To eliminate or at least drastically reduce the number of in vivo experiments, it is necessary to use a special apparatus of mathematical modeling. Based on this, the mathematical modeling’s technique of the intermolecular interactions process of cell membrane molecules with toxicants and antidotes to them is proposed in the paper. The main idea of the work is to study the formation’s process of stable bonds of toxicants’ molecules and antidotes with molecules of the cell membrane components , by identifying the active centers of this interaction. It uses specially created algorithms for constructing the structure of a conglomerate of two molecules, analysis and evaluation of the formation of a hydrogen bond between them. For this purpose, systems analysis, quantum chemical calculations, and modular programming are used to calculate the properties of individual molecules and the conglomerate as a whole. All received information is stored in specially designed databases. For a more visual presentation of the results, an original scheme for displaying the signatures of blocked active centers of the cell membrane for the antidotes in question has been proposed. The method of computer modeling outlined in the article allows a targeted search for antidotes to a given toxicant by creating a list ranked by the degree of effectiveness of antidotes.

Keywords: active centers of intermolecular interaction, the signature of the active centers of the components of the cell membrane, identification of the cell membrane of various organs of living organisms, active centers of toxic effects blocked by the antidote.

Full text:
AzhmukhamedovZharkikh_1_20_1.pdf

THE PORTAL REMOTE REHABILITATION “NEURODOM”: IMPLEMENTATION AND TESTING


UDC 378.146 +004.031.42
DOI:10.26102/2310-6018/2020.28.1.017

T.N. Ivanilova, V.A. Semenov, I.V. Vasilenko, M.A. Dneprovskayay, S.Z. Mirbadiev, Y.A. Yurinsky, S.V. Prokopenko, S.A. Subocheva

Remote neurorehabilitation refers to the field of telemedicine and allows the restoration of impaired body functions as a result of spinal injuries, traumatic brain injuries, stroke, heart attack, multiple sclerosis, etc. at the third stage of medical rehabilitation. The portal is designed to create continuous rehabilitation at home with feedback from the doctor; timely adjustment of the individual trajectory of patient rehabilitation; monitoring the results of rehabilitation through online monitoring of the patient’s health using the video recording function built into the system. Performing a set of exercises with the help of information and communication technologies in a remote format will optimally increase the physical, mental and social potential of the patient, for better integration into society. The article describes and presents the results of testing the portal of remote neurorehabilitation: safety testing, functional testing for the roles (doctor, patient, administrator) of the site was checked using test cases; performance testing; for usability – testing, a survey was conducted on real patients and a percentage ratio was compiled; testing the mobile version using the adaptivator tool, which revealed the pros and cons of the mobile version. Testing of the portal was carried out on 20 real patients. Identified errors during testing of the remote portal are resolved. The portal is at the stage of implementation at the Professor Clinic of KrasGMU named after Professors V.F. War-Yasenetsky.

Keywords: remote neurorehabilitation, telemedicine, information system, site testing process, test case.

Full text:
IvanilovaSoavtors_1_20_1.pdf

DIAGNOSIS OF EARLY STAGES OF ATTENTION DISORDERS BASED ON HYBRID FUZZY DECISION RULES

UDC 615.47
DOI:10.26102/2310-6018/2019.27.4.031

A.V. Polyakov, S.N. Rodionova, N.L. Korzhuk, L.V. Starodubtseva


The work is devoted to improving the quality of differential diagnosis of early stages of cognitive impairment in terms of fuzzy description of the studied classes of States.To assess the functions of attention, a device developed at the Department of biomedical engineering of SUSU is used to determine such properties of attention as concentration, volume, selectivity, switchability, distributability and stability. As a mathematical apparatus is used, the hybrid methodology for the synthesis of fuzzy decision rules. The basic element of which is the function of belonging to the studied classes of States (norm, mild cognitive impairment, moderate cognitive impairment, initial clinical stage) with a basic variable determined by the scales of the selected properties of attention.The decision on classification is made by the maximum value of the analyzed membership functions.The obtained mathematical models allow to diagnose the early stages of violations of all the studied functions of attention. Expert confidence in the obtained mathematical models exceeds 0.8. If, together with the level of attention, additional indicators characterizing the functional reserve, levels of psycho-emotional stress and fatigue and energy imbalance of the Meridian structures of the body are used, confidence in the correct classification of the early stages of attention disorders reaches a value of 0.9, which allows us to recommend the results in practical psychology and medicine.

Keywords: the early stage of cognitive impairment of attention, and hybrid decision rules, membership functions, confidence in the correct classification.

Full text:
PolyakovSoavtors_4_19_1.pdf

STUDY ON METHOD FOR MANAGING LOCATION AND PARAMETERS OF FOCAL SPOT FOR NON-INVASIVE SURGERY

UDC 534.222
DOI:10.26102/2310-6018/2019.27.4.007

E.G. Dombrugova, N.N. Chernov


The paper is dedicated to approbation of method for managing location and parameters of focal spot by changing physical characteristics of matching layer. A method is proposed for determining the speed of sound in matching layer for adjusting focal spot by calculating layer thickness-averaged speed of sound in biological tissues on the path of ultrasonic waves. A description of the experimental setup and research methodology is given. The influence of changes in the physical parameters of the matching layer on the distribution of the pressure amplitude in the frontal section of the acoustic field generated by the ultrasonic emitter when the angle of its inclination to the interface varies is estimated. The experimental data are compared with the results of mathematical modeling performed with the same input parameters. The results of calculations of the acoustic field generated by a focusing antenna array in a layered inhomogeneous medium (simulation layers) are presented and the influence of changes in the physical characteristics of the matching layer on the spatial distribution of the acoustic pressure amplitude in the focal region is estimated. The study results showed the possibility of using a matching layer with calculated parameters to control the location of the focal spot in the direction of ultrasonic waves propagation. The developed method for correcting the location of the focal spot can be used both for antenna arrays and for single emitters having the shape of a spherical segment.

Keywords: layer thickness averaged speed of sound, experimental setup, focused ultrasound, matching layer, refraction of ultrasonic waves.

Full text:
DombrugovaChernov_4_19_1.pdf