Method Of Multivariate Discriminant Analysis Research Articles

A MATHEMATICAL MODEL FOR PREDICTING THE OUTCOME OF TREATMENT OF MULTIDRUD-RESISTANT TUBERCULOSIS

The aim: Predicting the effectiveness of treatment for MRI of the lungs by developing a mathematical model to predict treatment outcomes. Materials and methods: 84 patients with MRI of the lungs: group 1 (n = 56) – with signs of effective TB treatment at the end of the intensive phase; group 2 (n = 28) – patients with signs of ineffective treatment. We used the multivariate discriminant analysis method using the statistical environment STATISTICA 13. Results: During the discriminant analysis, the parameters of the clinical blood analysis (monocytes, stab leukocytes, erythrocytes) were selected, which were associated with high (r> 0.5) statistically significant correlations with the levels of MMP-9, TIMP-1, oxyproline and its fractions and aldosterone in the formation of the prognosis. The mathematical model allows, in the form of comparing the results of solving two linear equations and comparing their results, to predict the outcome of treatment: “1” effective treatment, “2” – ineffective treatment. Early prediction of treatment effectiveness is promising, as it allows the use of the developed mathematical model as an additional criterion for the selection of patients for whom surgical treatment is recommended, in order to increase the effectiveness of treatment. Conclusions: An additional criterion for predicting ineffective MRI treatment, along with the criteria provided for by WHO recommendations, is a mathematical model that takes into account probably strong correlation (r = 0.5, p <0.05) between the factors of connective tissue destruction, collagen destruction, aldosterone , and indicators of a clinical blood test (between levels of OBZ and monocytes (r = 0.82, p = 0.00001), OB and monocytes (r = 0.92, p = 0.000001) OB and stab leukocytes (r = – 0.87, p = 0.0003) OBZ and stab leukocytes (r = – 0.53, p = 0.017), aldosterone and ESR.

Mathematical model for predicting treatment outcome in patients with destructive pulmonary multiresistant tuberculosis

The cessation of bacterial excretion is the main condition for the effective treatment of multiresistant tuberculosis (MRTB). Surgical procedures for treating tuberculosis patients significantly increase the effectiveness of treatment for such patients. The selection of new selection criteria for surgical treatment of MRI is relevant, as it will increase the effectiveness of MRI treatment, which remains insufficient.Objective — to predicting the effectiveness of treatment for MRI of the lungs by developing a mathematical model to predict treatment outcomes.Materials and methods. 84 patients with MRI of the lungs: group 1 (n = 56) — with signs of effective TB treatment at the end of the intensive phase; group 2 (n = 28) — patients with signs of ineffective treatment. We used the multivariate discriminant analysis method using the statistical environment Statistica 13.Results and discussion. During the discriminant analysis, the parameters of the clinical blood analysis (monocytes, stab leukocytes, erythrocytes) were selected, which were associated with high (r > 0.5) statistically significant correlations with the levels of MMP-9, TIMP-1, oxyproline and its fractions and aldosterone in the formation of the prognosis. The mathematical model allows, in the form of comparing the results of solving two linear equations and comparing their results, to predict the outcome of treatment: 1 — effective treatment, 2 — ineffective treatment. Conсlusions. Early prediction of treatment effectiveness is promising, as it allows the use of the developed mathematical model as an additional criterion for the selection of patients for whom surgical treatment is recommended, in order to increase the effectiveness of treatment.

State regulation of bankruptcy relations in the national economy

Introduction The deepening crisis in numerous business entities of various sizes is due to macroeconomic and domestic political instability, complicated by the consumer credit crisis, as well as tax and administrative pressure. Therefore, the acute problem of bankruptcy of business entities of different sizes and the need to regulate this issue at the state level. Unprofitable business entities of various sizes cause negative consequences for the domestic economy.
 Aim and tasks. The purpose of the work is to study state regulation of relations in the field of bankruptcy in the economy of Ukraine.
 Results. The factors, internal and external, preceding the bankruptcy of economic entities. Thus, the external preconditions of bankruptcy are international, economic, political, demographic and others. Internal prerequisites include economic, technical, technological, social and others. The characteristic and gradation criteria of business entities in accordance with legislative acts of Ukraine are presented. It was revealed that the proportion of sales of micro-enterprises is unequal to their share in the total number of enterprises, indicating their negligible contribution to the development of the national economy. According to the results of empirical studies, it was found that during the analyzed period the share of unprofitable enterprises of different sizes is at least a 25%. Such a large number entails not only low financial stability, insolvency and efficiency of business entities themselves, but also inhibits the process of economic growth of the whole country. The application of the method of multivariate discriminant analysis will allow us to draw a conclusion about the financial condition with assigning it a class in accordance with the values of the integral indicator and taking into account the probability of non-fulfillment of obligations, that is, to identify the presence of an unprofitable enterprise and the likelihood of bankruptcy.
 Conclusions. So, an analysis of the criteria for graduation of business entities made it possible to establish the difference between the administrative and accounting interpretation of the criteria for the attitude of enterprises to various sizes. The dynamics of structural statistics indicators of domestic business entities were monitored, suggesting that individual entrepreneurs make an insignificant contribution to the development of the national economy. Since a quarter of business entities still remain unprofitable, it is necessary to take preventive measures to prevent the loss of solvency by enterprises, one of which is to identify the existence of an unprofitable enterprise and the likelihood of bankruptcy.

Patterns in Style Evolution of Poets

This article investigates the problem of style stability and variation over time and presents the results of analysis of style evolution of American romantic poet H.W. Longfellow. The feature set includes 29 morphological and syntactic characteristics. The method of multivariate discriminant analysis is used. The study showed significant style changes over time and made it possible to single out the major tendencies of style evolution: the poet in his mature verses returns to the characteristics of his early lyrics. Besides, comparison of the variable and stable characteristics demonstrated that the initial part of the line is connected with style evolution, whereas the end of the line (which is the rhymed position) is less variable over time. The results may be useful for authorship attribution and manuscript dating research.

Prediction of Default of Small Companies in the Slovak Republic

Abstract From the time of Altman and the first bankruptcy prediction models, the prediction of default of companies is in the centre of interest of many economists and scientists all over the world. For companies, early detection of the possible threat of imminent financial difficulties or even bankruptcy is a very important part of financial analysis. Over the last few years, many predictive models have been created in the world. However, it has been shown that these models are not very well transferable to the conditions of the economy of another country and their prediction or rating power in another country is lower. Therefore, it is best to create a specific predictive model in the country that takes into account the situation of companies on the basis of real data on their financial situation. This paper is focused on creating a model of failure prediction of small companies in Slovakia using a well-known and widely used method of multivariate discriminant analysis. Discriminant analysis is one of the oldest multivariate statistical methods and sometimes it is difficult to fulfil certain assumptions for data. However, its results are easily interpretable and can be used to classify a company to the group of companies with risk of financial difficulties or, on the contrary, between well-prosperous companies. Prediction model is created based on real data on Slovak enterprises and has a strong classification ability in the specific conditions of the Slovak Republic.

Diffusion Capacity of the Lung and Maladaptive Post-Infarction Remodeling of the Heart

To assess the association of the lowering of diffusion capacity of the lung and maladaptive remodeling of the heart in a year after myocardial infarction (MI). We included into this study 107 patients with ST elevation MI (STEMI). Examination of all patients on 7-12 th days included echocardiography (echoCG), spirography, body plethysmography, and determination of lung diffusion capacity. After one year of observation in 87 patients (81.3%) we performed echocardiogram in dynamics. After 1 year signs of maladaptive remodeling were detected in 25 patients (28.7%). The group of patients with remodeling had clinically significantly lowered parameters of lung diffusion function: diffusion capacity corrected for hemoglobin (DLCOcor) by 30% (p=0.001) and diffusing capacity for carbon monoxide/alveolar volume (DL/VA) by 21% (p=0.001). A mathematical model for prediction of maladaptive cardiac remodeling was obtained by the method of multivariate discriminant analysis. Univariate analysis showed that the reduction of DLCOcor was associated with 3.5-fold elevation of the risk of maladaptive remodeling in one year after STEMI. After adjustment for risk factors this risk rose up to 13.7-fold. Reduction of lung diffusion capacity below 80% of the predicted value on the 12-th day of STEMI was associated with 13.7-fold increase of the risk of detection of maladaptive remodeling of the heart after one year. Independent predictors of cardiac maladaptive remodeling were the number of diseased coronary arteries, reduced DLCO and the presence of myocardial remodeling during acute phases of myocardial infarction.

Artificial Neural Network applied as a methodology of mosquito species identification

There are about 200 species of mosquitoes (Culicidae) known to be vectors of pathogens that cause diseases in humans. Correct identification of mosquito species is an essential step in the development of effective control strategies for these diseases; recognizing the vectors of pathogens is integral to understanding transmission. Unfortunately, taxonomic identification of mosquitoes is a laborious task, which requires trained experts, and it is jeopardized by the high variability of morphological and molecular characters found within the Culicidae family. In this context, the development of an automatized species identification method would be a valuable and more accessible resource to non-taxonomist and health professionals. In this work, an artificial neural network (ANN) technique was proposed for the identification and classification of 17 species of the genera Anopheles, Aedes, and Culex, based on wing shape characters. We tested the hypothesis that classification using ANN is better than traditional classification by discriminant analysis (DA). Thirty-two wing shape principal components were used as input to a Multilayer Perceptron Classification ANN. The obtained ANN correctly identified species with accuracy rates ranging from 85.70% to 100%, and classified species more efficiently than did the traditional method of multivariate discriminant analysis. The results highlight the power of ANNs to diagnose mosquito species and to partly automatize taxonomic identification. These findings also support the hypothesis that wing venation patterns are species-specific, and thus should be included in taxonomic keys.