Русский
06.2015 - Volume 19 Issue 2 (PDF)
Unlike CO2, with successive breath holds, with increasing time, the concentration of carbon monoxide (CO) tends to a certain equilibrium value (equilibrium concentration), depending on the content of carboxyhemoglobin in the blood. It is shown that the diffusion capacity of the lungs calculated by the rate of increase in CO concentration in the alveolar volume is approximately twice as large as their total diffusion capacity determined by the well-known "single breath" method, i.e. it corresponds to the diffusion capacity of the alveolar-capillary membrane. Based on mathematical modeling of the process of CO gas exchange in human lungs, a formula was obtained for calculating the integral measured concentration of CO in exhaled air, taking into account the effect of the dead (anatomical and instrumental) volume on the measurement results. The effect of the dead volume was taken into account under the assumption that gas exchange does not occur in it, but the CO contained in the inhaled air first enters the exhaled volume from it, and then the exhaled part of the alveolar CO. It was also taken into account that at the beginning of inhalation the dead volume is filled with the final portion of alveolar air that entered it as a result of the previous exhalation. Two models are considered: the linear model (LM), when CO gas exchange in the changing alveolar volume occurs at a constant transfer coefficient, and the elastic shell model (ESM), when the transfer coefficient during breathing changes proportionally to the membrane surface area and inversely proportional to its thickness. As a result of the analysis of the obtained results, it is shown that calculations in the ESM approximation more adequately describe the experimental data and can be used to solve the inverse problem, that is, calculating the diffusion capacity of the alveolar-capillary membrane, the alveolar volume of the lungs and the equilibrium concentration of CO, based on three values of the measured concentration of CO in different breathing modes. Thus, the proposed method, unlike "single breath", allows determining the indicated important physiological parameters without using test gas mixtures.
Keywords: gas exchange, carbon monoxide, diffusing capacity of the lungs, alveolar volume, mathematical modeling.
The paper presents the formulation of the problem, within the framework of which the Medical Protocol Description Language (MPDL) was developed, describes this language, which is essentially a specialized high-level programming language, and lists the areas in which MPDL is in demand.
Keywords: specialized high-level programming language, medical decision support systems.
The paper will prove sufficient conditions for the continuity of the lattice of propositional calculuses with the substitution operation and arbitrary circuit operations of inference. In addition, the cardinalities of the sets of all closed classes, precomplete and dual-precomplete classes of tautologies in the lattice of propositional calculuses will be described. The types of criteria systems for propositional calculuses will also be described.
Keywords: lattice of propositional calculi, precomplete classes, dual-precomplete classes, criteria systems.
This paper presents the results of an experimental analysis of the properties of Abelian groups in connection with their cryptographic applications. Due to the complexity of the computational problem, groups of small order are considered.
Keywords: order of 2-transitivity, Abelian group, finite group, symmetric cipher.
This paper considers shape models of objects in an image: binary and multi-class Boltzmann models. A new algorithm for training a multi-class Boltzmann shape model is proposed, for the application of which it is sufficient to have incomplete data labeling, namely: binary labeling and specifying seeds indicating the approximate location of object parts.
Keywords: Boltzmann shape model, multi-class Boltzmann shape model, graphical models, EM algorithm.
This article describes the process of creating a web application that implements the functionality of a geographic map. The material is divided into two parts. The first part solves the problems of drawing a map on the screen, scaling the map, moving around the map, displaying objects on the map. The second part considers the problem of clustering a predetermined set of objects located on the map using the k-means method and the distance method.
Keywords: online map, single-page applications, tiles, php, jQuery, ajax, SQL, k-means method, distance method, clustering.
The paper considers one-variable iterative propositional calculi, which are finite sets of propositional formulas in one variable together with the modus ponens operation and the superposition operation defined by the set of Mal'tsev operations. For such calculi, the problem of derivability of formulas will be reduced to the problem of deriving words in linear canonical systems. In particular, it will be shown that all one-variable iterative propositional calculi are decidable.
Keywords: iterative propositional calculi, derivability problem, linear canonical systems.
The article studies the relationship between the membership of automaton internal functions in precomplete classes of Boolean functions and the automaton's implementation of a dictionary function from the same class.
Keywords: automata theory, precomplete classes, dictionary function.
The article considers two problems: the problem of embedding admissible classes of alphabetic coding and the problem of embedding admissible classes of regular languages. In the first case, for an arbitrary pair of regular languages in a common alphabet, it is necessary to understand whether it is true that any alphabetic coding that is bijective in the first language will also be bijective in the second. In the second case, for an arbitrary pair of alphabetic codings in common input and output alphabets, it is necessary to understand whether it is true that an arbitrary regular language in which the first coding is bijective will have the same property in the second coding. It is shown that the first problem is algorithmically solvable for the case when the cardinality of the input alphabet is two. In the second case, it is shown that the problem is always algorithmically solvable.
Keywords: alphabetic coding, regular languages, embedding problem, admissible classes.
The paper formulates and proves a criterion for the realizability of functions on strings by probabilistic Moore automata with numerical output.
Keywords: probabilistic automata, Moore automata, reactions, random functions.
The article considers languages in the alphabet \(\{a_1, . . . , a_n\}\), in whose words the share of all consecutive pairs \(a_ia_j\) is fixed. This share is described by the generating matrix of the language Θ. The author called such languages bigram languages. Natural languages have a similar property. It turns out that the properties of such languages to be empty, finite, regular, context-free or context-dependent are verified by the matrix Θ. In this paper, the issue of infinite context-free languages is considered in detail.
Keywords: bigram, bigram multiplicity matrix, bigram language, context-free language, directed graph, Euler graph.
The paper constructs a generalization of the non-influence automaton model for the case of quantum automata. A "spin-up theorem" is proved, which describes sufficient security conditions.
Keywords: finite automata, system security, non-influence model.