Abstracts
Збережено в:
| Дата: | 2014 |
|---|---|
| Формат: | Стаття |
| Мова: | English |
| Опубліковано: |
Міжнародний науково-навчальний центр інформаційних технологій і систем НАН України та МОН України
2014
|
| Назва видання: | Кибернетика и вычислительная техника |
| Онлайн доступ: | http://dspace.nbuv.gov.ua/handle/123456789/84508 |
| Теги: |
Додати тег
Немає тегів, Будьте першим, хто поставить тег для цього запису!
|
| Назва журналу: | Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| Цитувати: | Abstracts // Кибернетика и вычислительная техника. — 2014. — Вип. 175. — С. 88-94. — англ.. |
Репозитарії
Digital Library of Periodicals of National Academy of Sciences of Ukraine| id |
irk-123456789-84508 |
|---|---|
| record_format |
dspace |
| spelling |
irk-123456789-845082015-07-10T03:02:08Z Abstracts 2014 Article Abstracts // Кибернетика и вычислительная техника. — 2014. — Вип. 175. — С. 88-94. — англ.. 0452-9910 http://dspace.nbuv.gov.ua/handle/123456789/84508 en Кибернетика и вычислительная техника Міжнародний науково-навчальний центр інформаційних технологій і систем НАН України та МОН України |
| institution |
Digital Library of Periodicals of National Academy of Sciences of Ukraine |
| collection |
DSpace DC |
| language |
English |
| format |
Article |
| title |
Abstracts |
| spellingShingle |
Abstracts Кибернетика и вычислительная техника |
| title_short |
Abstracts |
| title_full |
Abstracts |
| title_fullStr |
Abstracts |
| title_full_unstemmed |
Abstracts |
| title_sort |
abstracts |
| publisher |
Міжнародний науково-навчальний центр інформаційних технологій і систем НАН України та МОН України |
| publishDate |
2014 |
| url |
http://dspace.nbuv.gov.ua/handle/123456789/84508 |
| citation_txt |
Abstracts // Кибернетика и вычислительная техника. — 2014. — Вип. 175. — С. 88-94. — англ.. |
| series |
Кибернетика и вычислительная техника |
| first_indexed |
2025-07-06T11:30:16Z |
| last_indexed |
2025-07-06T11:30:16Z |
| _version_ |
1836896924632875008 |
| fulltext |
88
ABSTRACTS
L.S. Fainzilberg, E.N. Minina
EVALUATION OF THE FUNCTIONAL STATE OF THE
CARDIOVASCULAR SYSTEM USING MAGNITUDE VARIATION OF
THE PHASE TRAJECTORIES OF SINGLE-CHANNEL ECG
Introduction: The new approach to the assessment of the functional state of
the cardiovascular system during screening examinations based on the analysis of
the phase portrait’s features of a single-channel ECG is developed. Our previous
studies have shown that using of new diagnostic features of the phase portrait
single channel ECG is possible to increase the sensitivity and specificity of
diagnosis even in cases where traditional analysis of 12-lead ECG is
uninformative. We assume that the degree of dispersion of the phase portrait
trajectories of single-channel ECG may incur an additional diagnostic value as a
predictor of the adequacy of the "degree of randomness" of the functioning of the
system with regard to its level of development, gender and individual
characteristics of the organism.
The purpose of the article is experimental verification of this hypothesis on
clinical material.
Methods: Studies were conducted using a hardware-software complex
FASEGRAPH that implements information technology processing of the first
standard lead of ECG on the phase plane. In contrast to well-known works, in
which a single-channel ECG phase portrait is studied on a plane )(),( τ−tztz ,
where const=τ is a time delay, the complex constructs the phase portrait in the
coordinates )(),( tztz & where )(tz& is the rate of change of the signal of the electrical
activity of the heart. Due to this we take into account not only the amplitude but the
speed characteristics of the test process. We investigate the diagnostic value of
indicator σ which characterizes the dispersion of ECG’s trajectories obtained from
129 female and 192 man from 6 to 76 years old. Our goal was to study the
dependence of the indicator of age and gender. A secondary objective was to
compare values of indicator σ with the level of fitness.
Results: It was established that with increasing age the indicator σ tends to
increase among women and to decrease among men. Indicator's changes relative to
the average values was as high as 15% ( 05,0<p ). Children under 12 regardless of
gender demonstrate a wide range of changes of the indicator which was an average
22.5% ( 05,0<p ) higher than the rate of an adult. By increasing the external load
among male athletes the increase in the index σ exceeds the increase in heart rate.
The opposite trend is observed among woman-athletes. By increasing the external
load in persons who are not involved in sports, regardless of their sex an increasing
of the heart rate exceeds the increase in the indicator σ . Among athletes the
indicator σ changes in waves under increasing the load while in the non-sporting
group the indicator σ increases monotonically. Quantitative changes of the
indicator σ under urgent and long-term adaptation is directly related to the optimal
switching high-frequency components of heart rate regulation.
Conclusions: The dispersion of single-channel ECG’s trajectories in the phase
space obtained by measuring of the indicator σ can be used as an additional
criterion for evaluating the effectiveness of the management the cardiovascular
system’s functional state.
Key words: information technology, the phase portrait of ECG, criterion.
89
M.I. Vovk, Ye.B. Galyan, E.N. Podoprigora
INFORMATION TECHNOLOGY FOR MOVEMENT CONTROL OF
THE HAND USED TO RESTORE THE MOTOR COMPONENT OF
SPEECH
Introduction: Cerebral strokes not only lead to disorders of the motor
functions, but also, in 50–60% of cases, of the speech. In our previous studies we
developed structural and functional model of the speech motor component
restoration based on connecting an external circuit of the purposeful control of
forced muscle movements of the hand to the patient. The central part of this circuit
is a programmed electrical muscle stimulator. Further studies are aimed at solving
practical problems of technology synthesis of speech restoration.
The purpose of this research is to develop an information technology of
purposeful control of hand movements to restore the motor component of speech as
well as to conduct clinical trials of the technology.
Method: The technology under consideration is based on programmed
electrical muscle stimulation method.
Results: Mechanisms of neural plasticity, link between speech development
and the development of finger movements as well as proximity of centers of speech
and hand representation in the cerebral cortex, all serve as the theoretical
grounding for the synthesis of information technology of movement control of the
hand to restore speech.
Training programs serving as data signals to control the movements of the
hand, methods of transmission of these signals, types of fine motor hand
movements, and topology of transmitting the signals to the muscles to control these
movements, were all considered as components of this technology. The structural
and functional model for speech recovery based on the information technology of
purposeful control of fine motor hand movements was presented.
Training of fine hand movements based on the electrical stimulation driven by
artificially synthesized programs leads to the formation of new motor centers for
the hand and speech. The “Donor” program (the mirroring program), in which the
nonparalyzed hand is fit with a sensor to record voluntary fine motor hand
movements, and the opposite affected hand is fit with surface electrodes to active
the paralyzed muscles, enables to mobilize additional mechanisms of neural
plasticity. The criteria for selecting fine hand movements used in speech
restoration encompassed development of fine motor skills of the hand in ontogeny
and availability of motor points of muscles during electrical stimulation with
surface electrodes.
The results of clinical trials of the technology in the complex of rehabilitation
measures used to restore speech in post-stroke patients with motor aphasia and its
analysis show improvement in language function in 63% of patients, compared to
improvement rates of 40%, when the traditional course of rehabilitation is
followed.
Conclusions: Use of information technology in purposeful control of fine
hand movements in the rehabilitation of stroke patients contributes to the
restoration not only of hand movements but also of speech.
Keywords: information technology, purposeful control, programmed
stimulation, fine hand movements, structural and functional model, speech
restoration, neural plasticity, clinical trials.
90
S.I. Dotsenko, P.A. Negadailov
ON OPTIMAL SEARCH ORDER IN THE GROUP SECRETARY
PROBLEM
Introduction: We consider a problem of choosing the best item for the case
when the elements are divided into groups and performed in a single step
simultaneous viewing of elements of the group.
A case, when the objects are divided into groups and carried out simultaneous
viewing of candidates in each group, is considered in [3]. After watching the group
of candidates the decision procedure is similar to the classical problem, namely: if
the group presents the best candidate among all previously viewed items (such an
element is called a maximum) to make a decision to choose this candidate and
finish viewing or reject it and continue, the returning to the previously rejected
candidates is prohibited.
In this case the optimal rule for selecting the best candidate based on the so
called “choice theorem” (or “Bruce’s theorem”) is considered in [4] because
maximum elements arising may be interpreted as independent Bernoulli trials with
different probabilities of success. In this connection a natural question arises: in
what order one should view groups that maximize the probability of selecting the
best candidate.
Purpose: We try to find the best order of viewing groups which maximize the
probability of selecting the best candidate, provided that optimal stopping rule,
based on the “Bruce’s theorem” is applied and we compare this probability for the
best and the worst cases. As may be expected, the lower bound for the worst case is
the probability to find the best element at the classical secretary problem, i.e. 1/e.
Results: At first we prove two lemmas concerning the form of the optimal
order of viewing groups, allowing to understand the structure of optimal solutions.
The knowledge of structure helps to shrink the search to some class of
permutations (considerably more narrow, then the initial set) of groups. Then we
constructed genetic algorithm which finds the approximate optimal solution.
Conclusion: The sequence of groups that maximize the probability of
successful choice at the generalized secretary problem must follow the certain
rules, that are formulated as lemma 1 and 2. Nevertheless, permutation number that
are follow the mentioned rules remains big enough and it impossible to consider all
of them directly.
The genetic algorithm finds the approximate solution numerically with high
precision. For the case of three groups the exact rule and correspondent probability
of success are obtained.
Keywords: optimal choice, secretary problem, Bernoulli trials, group search,
genetic algorithm.
91
A.V. Gummel
STUDY OF A CONTROL PROCESS SYNTHESISED USING
APPROXIMATE MODEL WHICH IS IDENTIFIED UNDER BOUNDED
ERROR CONDITIONS
Introduction: Majority of existing control methods consider that
mathematical model of controlled object is given a priori. If the model is unknown,
then identification methods are applied. This methods allows to retrieve a model of
researched object using an input and output data which obtained from conducted
on object experiments. In case when experiment information contain stochastic
white noise bias model structure and parameters can be recovered in a way that
accuracy of identification is sufficient to apply existing methods of robust synthesis
and obtain acceptable control to get required properties of closed loop system. If
only known information is only some boundary of an errors full recovery in many
cases is impossible. This study deals with the problem of control synthesis using
model obtained by system identification from information affected by bounded
error. Such models with given identification conditions can only represent a
reduced dimension approximation of original system. Previous research managed
to develop identification methods allowing to get approximation models, precision
of which is consistent with bias in original data. For linear stationary systems such
methods consist of two essential stages. On first stage model structure is
determined using singular values as result of SVD procedure applied to input and
output information matrices. Second stage is to determine parameters of model
given structure of model.
The purpose of this paper is to determine to what extent identified
approximating models are suitable for solution of practical problems, how this
models should be used in control synthesis problems and ultimately offer a
reasonable method of solving control synthesis problem using identified model.
Results: Paper describes identification process and following control
synthesis to assure target properties of system with feedback. The modified
4SID-method is used for indentification. Feedback control is synthesized using
poles positioning principle and asymptotic state identification of system. Data is
obtained using numerical modeling allow to make some evaluation of synthesized
control quality.
Conclusions: Based on the results obtained during the research it is possible
to state that identified approximating models can be used to change the dynamic
characteristics of original unknown system.
Properties of synthesized control can allow to get a system with feedback that
has shorter settling time and keep required level of output signal. Whereas bias in
system with feedback will stay within error boundaries of initial system or at least
overshoot will be insignificant. However further decrease of settling time will lead
to increase of bias level of system with feedback, thus such synthesis should be
considered as a trade-off between acceptable bias level and required speed of
transient response.
Keywords: system identification, control synthesis, 4SID, asymptotic state
identification.
92
V.V. Pavlov, Y.M. Shepetukha
ISSUES OF NET-CENTRIC SYSTEMS ANALYSIS AND SYNTHESIS
Introduction: New scientific endeavors connected with the creation of net-
centric concept as well as the development of net-centric systems are widely used
in different fields of human activity. Advanced computer and communication
technologies allow a real-time modeling of distributed functional and operational
components for complex applied processes.
The purpose is the elaboration of methodological issues related to analysis as
well as synthesis of net-centric systems with high dimensions and complex
configurations.
Methods: General system theory, conflict resolution theory, invariant systems
synthesis, theory of deterministic chaos, methods of chaos synchronization and
control.
Results: In addition to the traditional taxonomy, specifying man-machine
systems operating under the conditions of certainty, risk or uncertainty, it is
proposed to distinguish systems working within the so-called “deterministic chaos”
when small variations of initial conditions generate completely different phase
trajectories. Methodology for the development of net-centric structures should be
based upon a system integration of interacting functional and operational
components. Modeling of applied processes allows to examine possible scenarios
that describe originating of system’s chaotic behavior. The proposed approach
could also be used for both quantitative and qualitative analyses of essential
parameters’ influence on different options for a recovery of net-centric systems’
organized state.
Conclusions: An important issue of research in the field of net-centric
systems is recognition and structuring of specific functional and operational
elements coping with internal as well as external information flows. Examination
of applied processes in distributed network structures gives a possibility to
distinguish various models of chaotic behavior: discrete models, finite phase space
ones described by ordinary differential equations (systems of Lorenz, Rossler,
etc.), infinite phase space models governed by ordinary differential equations with
delays (system of Ikeda, etc.). Another significant problem of net-centric systems’
construction is the examination of a way from chaotic to organized state of the
system. Such a backward transition could be performed with the use of
modification for essential variables of information exchange process.
Keywords: applied process, net-centric system, chaotic behavior, chaos
control, non-linearity, unstable phase trajectory, synchronization, feedback.
93
A.A. Trizna, S.A. Yankova
AREAS OF PRACTICAL APPLICATION OF INDICES OF
AIRCRAFTS MAINTENANCE AND THEIR GROUND BY METHOD OF
THE GENERALIZED ENTROPIES
Introduction: There is a necessity of classification and specification of
determinations of aircraft maintenance indexes and criteria.
In theory, the technical operation is now entering its second phase. Its
generalized characteristic is to consider not operating as an independent phase, and
as the central stage in the life cycle of the existence of industrial products, the need
for advanced classification and clarify the definitions and criteria of technical
manuals; generalized analytics system properties (efficiency, reliability,
maintainability, durability, cost and etc.) and use these combinations of properties.
The purpose of the article is aircraft maintenance and engineering psychology
indexes as complex poliсomponent probabilistic values of these indexes and
determination of area of their practical applicability.
Results: Regularity researches of additive and multiplicative measures show
that at the increase of number of components in an index the area of practical
applicability of additive index passes to the area of small probabilities and
multiplicative — to the area of high probabilities.
The developed table with nomograms for the engineering calculation of
practical application of complex poliсomponent probabilistic values domains
allows to define a practical application of these indexes domain at the estimation of
aircraft maintenance.
Conclusions: Theory and practice of technical operation of the aircraft is
currently at the stage of organizing and summarizing data. In order to organize, you
need to generalize the use of such approaches as a process approach, process
analysis and generalized theory of processes. Analysis of indicators of aircraft
operations showed that the calculation of areas of their practical applicability
should be carried out by the generalized entropy process. Developed classification
performance of aircraft operation, which is based on the consideration of these
indicators as multicomponent probability values (measures) to determine the area
of practical application of these indicators when assessing the technical operation
of aircraft.
Investigation of the laws of the additive and multiplicative measures showed
that increasing the number of components in the index, the area of practical
applicability of the additive indicator moves to the low-probability, and
multiplicative function in the region of high probability.
Keywords: Aircrafts maintenance, engineering analysis, indices of
maintenance, entropy, nomogram.
94
Kochina M.L., Nessonova М.N.
METHODS TO ESTIMATE PATIENT’S GRADE OF SEVERITY
Introduction: One of the actual tasks in developing medical decision support
systems is forming of decision rules, which serve for patient’s grade of severity
estimation. Since medical data are mostly random variables, the most appropriate
techniques for mathematical modeling in medical applications are stochastic,
probabilistic and statistical methods of multivariate analysis.
The purpose of the paper is to develop the methods for patient’s grade of
severity estimation, which involve the method to construct classifiers, and methods
to form their compositions.
The method for classifiers constructing is based on the metric approach to
graphical representations (maps or plots) generated by statistical methods of
geometrical representation of data structure (multidimensional scaling and
correspondence analysis). The key point of the method developing is that we
represent “classes” not as some subspaces or areas of compact layout in objects’
space, but as an additional feature of object’s description. Using such interpretation
of “class” notion we construct the mathematical model of target feature “class”
dependence on other explanatory variables.
To construct compositions of classifiers we develop and formalize the method,
which we call “rate voting”. The method uses weights of basic classifiers,
evaluated as functions of basic algorithms’ answers for an object, their accuracies
in predicting separate classes, and mistakes in prediction other classes. Another
method of classifiers compositions constructing were called “seniority rate voting”.
The method evolves ideas of “rate voting” by combining them with principles of
committees with seniority logic. This combined method turns to be more relevant
and effective in some practical applications.
Results and Discussion: The applying results of the methods suggested are
reported in the paper. The mathematical models for patient’s grade of severity
estimation in a case of pancreas trauma and traumatic pancreatitis, for differential
diagnostics of bile-excreting ducts diseases, for clinical outcome prediction in a
case of cerebral strokes and pancreas traumatic injuries, were developed. It is
shown that overall accuracy, sensitivity and specificity of these models are greater
than of those, which were developed on the basis of the common mathematical-
statistical modeling methods.
Conclusions: The methods to construct classifiers and their compositions,
which are reported in the paper, allow improving of overall accuracy, specificity
and sensibility of patient’s grade of severity estimation.
Keywords: patient’s grade of severity estimation, supervised classification
method, compositions (committees, ensembles) of classifiers.
|