Wyniki wyszukiwania - BazTech

1

The pair-wise linear classifier and the k-NN rule in application to ALS progression differentiation

Sokołowska B., Jóźwik A., Niebroj-Dobosz I., Janik P.

Journal of Medical Informatics & Technologies

|

2012

|

Vol. 20

79--83

EN

The two kinds of classifier based on the k-NN rule, the standard and the parallel version, were used for recognition of severity of ALS disease. In case of the second classifier version, feature selection was done separately for each pair of classes. The error rate, estimated by the leave one out method, was used as a criterion as for determination the optimum values of k's as well as for feature selection. All features selected in this manner were used in the standard and in the parallel classifier based on k-NN rule. Furthermore, only for the verification purpose, the linear classifier was applied. For this kind of classifier the error rates were calculated by use the training set also as a testing one. The linear classifier was trained by the error correction algorithm with a modified stop condition. The data set concerned with the healthy subjects and patients with amyotrophic lateral sclerosis (ALS). The set of several biomarkers such as erythropoietin, matrix metalloproteinases and their tissue inhibitors measured in serum and cerebrospinal fluid (CSF) were treated as features. It was shown that CSF biomarkers were very sensitive for the ALS progress.

2

Pattern recognition approach for analysis of metabolic response to intermittent hypoxia

Sokołowska B., Jóźwik A.

Journal of Medical Informatics & Technologies

|

2010

|

Vol. 15

177--183

EN

Intermittent hypoxia (IH) elicits two forms of respiratory plasticity, which are initiated during and after exposure to IH, i.e. a long-term facilitation and a progressive augmentation of respiratory motor output. IH is often used as a model of sleep apnea and/or respiratory plasticity in humans and animals. Procedures of IH are also applied in sport medicine and rehabilitation of respiratory diseases. The aim of the present paper is an analysis of a metabolic response to acute intermittent hypoxia in a rat model. The animals were placed and monitored in a whole body plethysmographic chamber. The rats were exposed to five consecutive cycles consisting of 10-min hypoxic stimulus period separated by 10-min normoxic intervals, and additionally they were monitored up to 1 h after the final hypoxic exposure. The metabolism software analyzer recorded following variables (features): metabolic rate, carbon dioxide production, oxygen consumption and respiratory quotient. The obtained results demonstrated that acute IH causes metabolic effects during and after intermittent stimuli, which may be effectively recognized by an application of the k-NN classifiers.

3

Recognition of pathological states in arterial blood by distance based techniques

Sokołowska B., Jóźwik A.

Journal of Medical Informatics & Technologies

|

2003

|

Vol. 5

MI23--30

EN

The paper presents the application of some distance based pattern recognition algorithms for recognition of pathological states in respiratory system on the basis of the arterial blood gasometry (features pH, pCO2, pO2). In our biological model two experimental situations were considered: 1) the intact animals and 2) the main inspiratory muscles paralyzed (after acute of bilateral phrenicotomy). The comparison of the mentioned three features in the two conditions was the main goal of the present study. The analyzed biological data set contained 38 in class 1 (muscle function preserved) and 36 in class 2 (after diaphragm paralyzed) measurements. It was discovered that a significant part of the measurements could be correctly recognized as the ones coming from the first or the second class according to gasometric measurements.

4

Distance based classifiers and their use to analysis of data concerned acute coronary syndromes

Jóźwik A., Stawska Z., Grabowski M., Filipiak K., Rudowski R., Opolski G.

Image Processing & Communications

|

2003

|

Vol. 9, no 3-4

45-52

EN

The paper presents effectiveness of classifiers based on distance function in application to real problem concerned acute coronary syndromes. The types of decision rules: the standard k-NN rule; its fuzzy version and the multistage decision rule that uses the class overlap idea are considered. In the case of the fuzzy k-NN rule the fuzzyness is applied only for decreasing a misclassification rate. The multistage classifier is taken into account because of its very desired property, which consist in possibility of determination whether a case being classified is difficult or easy for recognition. The more difficult is the case to be classified the more stages are required. This property enables an error rate gradation. In each stage the proposed classifier can make up one of the three following decisions: indicate a class number, reply :"I do not know" or qualifythe object to the next stage. A number of stages depend on the classified object. The analyzed data concern to the two-class decision problem that consist in prediction whether the patient will survive the period of one month or not.

5

Some problems with construction of the k-NN classifier for recognition of an experimental respiration pathology

Jóźwik A., Sokołowska B.

Journal of Medical Informatics & Technologies

|

2002

|

Vol. 3

MI89--97

EN

An objective of the work is to demonstrate some difficulties with construction of a classifier based on the k-NN rule. The standard k-NN classifier and the parallel k-NN classifier have been chosen as the two most powerful approaches. This kind of classifiers has been applied to automatic recognition of diaphragm paralysis degree. The classifier construction consists in determination of the number of nearest neighbors, selection of features and estimation of the classification quality. Three classes of muscle pathology, including the control class, and five ventilatory parameters are taken into account. The data concern a model of the diaphragm pathology in a cat. The animals were forced to breathe in three different experimental situations: air, hypercapnic and hypoxic conditions. A separate classifier is constructed for each kind of the mentioned situations. The calculation of the misclassification rate is based on the leave one out and on the testing set method. Several computational experiments are suggested for the correct feature selection, the classifier type choice and the misclassification probability estimation.

6

Practical approaches to statistical pattern recognition

Jóźwik A.

Biocybernetics and Biomedical Engineering

|

2002

|

Vol. 22, no. 2-3

207-221

EN

The paper describes new approaches to statistical pattern recognition. All presented methods are based on a distance function. The properties of these methods and their usefulness are illustrated on real problems. Some tasks with small and very large training sets are described to shown an effectiveness of the proposed approaches. There is no one universal method that would be satisfactory for all object classification problems. That's why several methods have been demonstrated.