Wyniki wyszukiwania - BazTech

1

Multiclassifier systems applied to the computer-aided sequential medical diagnosis

Kurzyński M., Majak M., Żołnierek A.

Biocybernetics and Biomedical Engineering

|

2016

|

Vol. 36, no. 4

619--625

EN

The diagnosis of patient's state based on results of successive examinations is common task in the medicine. In computer-aided algorithms taking into account the patient's history in order to improve the quality of classification seems to be very reasonable solution. In this study, two original multiclassifier systems (MC) for the computer-aided sequential diagnosis are developed, which differ with decision scheme and the methods of combining of base classifiers. The first MC system is based on dynamic ensemble selection scheme and works in two-level structure. The second MC system in combining procedure uses original concept of meta-Bayes classifier and produces decision according to the Bayes rule. Both MC systems were practically applied to the diagnosis of human acid–base equilibrium states and compared with some state-of-the-art sequential diagnosis methods. Results obtained in experimental investigations imply that MC system is effective approach, which improves recognition accuracy in sequential diagnosis scheme.

2

On new methods of dynamic ensemble selection based on randomized reference classifier

Krysmann M., Kurzyński M.

Journal of Medical Informatics & Technologies

|

2012

|

Vol. 20

101--107

EN

In the paper two dynamic ensemble selection (DES) systems are proposed. Both systems are based on a probabilistic model and utilize the concept of Randomized Reference Classifier (RRC) to determine the competence function of base classifiers. In the first system in the selection procedure of base classifiers the dynamic threshold of competence is applied. In the second DES system, selected classifiers are combined using weighted majority voting rule with continuous-valued outputs, where the weights are equal to the class-dependent competences. The performance of proposed MCSs were tested and compared against DES system with better-than-random selection rule using eleven databases taken from the UCI Machine Learning Repository. The experimental results clearly show the effectiveness of the proposed methods.

3

Concept of a system for training of bioprosthetic hand control in one side handless humans using virtual reality and visual and sensory biofeedback

Wołczowski A., Kurzyński M., Zapłotny P.

Journal of Medical Informatics & Technologies

|

2011

|

Vol. 18

85--91

EN

In the paper the concept of a training system is presented which can help to stimulate sensory-motor cortex centers in order to develop their ability for efficient use of bioprosthesis. The basis of the training system is a virtual reality with a virtual hand, that the trained patient can move and concurrently observe the movement on the screen (visual feedback) and whose contact with virtual objects the patient may feel as a touch (sensory feedback). The construction of the virtual hand consists of physical elements, connected by joints, a graphical object representing the structure of the hand and the bones enable its deformation. The control procedure of virtual hand is realized through recognition of intention of hand motion on the basis of EMG signals coming from the stump muscles. The recognition algorithm is constructed using the learning set, i.e the set of pairs containing the class of hand fingers movement and accompanying myopotentials segments, which are acquired from the muscles of healthy hand.

4

Medical Telemetry System for Monitoring and Localization of Patients : Functional Model and Algorithms for Biosignals Processing

Kurzyński M., Ryba P., Markowski M., Woźniak M.

International Journal of Electronics and Telecommunications

|

2010

|

Vol. 56, No. 4

445-450

EN

The article presents the concept of hospital telemetric system. The goal of the project is a model of early warning systems for patients outside intensive care wards. Proposed system is based on constant telemetric monitoring using objective physiological parameters. Using low-distance sensor network which covers body of a patient, so-called BAN (Body Area Network) is the main innovation of the project. Some preliminary results of ECG analysis and interpretation modules and units of proposed system will be presented.

5

An improved diffusion driven watershed algorithm for image segmentation of cells

Tarnawski W., Kurzyński M.

Journal of Medical Informatics & Technologies

|

2007

|

Vol. 11

213--220

EN

The image segmentation is one of the most crucial steps in automated analysis of medical and biological images. The segmentation process allows for a detection of object contours. Due to specificity of imaging technique, a correct detection of cell contours is problematic because of the fuzzy and broken edges. Moreover, the cells are very often connected. The modified watershed algorithm based on the diffusion model presented in this paper has been successfully applied to segmentation of cells where the mentioned difficulties appear. The method was tested in contact endoscopy, a novel technique in the diagnosis of the larynx.

6

Computer system of telemedical services for family doctor's practices

Kurzyński M., Jóźwiak G., Moi M.

Journal of Medical Informatics & Technologies

|

2002

|

Vol. 3

MI99--107

EN

This paper shows the current state of research and development over the telemedical services system for family doctor's practices. The introduction of this telemedical system should result in improvement in accessibility to medical services for the patient and to actual medical information, publications and up-to-date consultations for the physician, regardless of the distance from physician's clinic to the consultation center. This system should reduce the total costs of basic medical services. The system consists of two autonomous subsystems, that work in two arrangements: patient ↔ family doctor's practice and family doctor's practice ↔ information & consultation center. The remote medical care system provides vital signs monitoring through telephone line - physician can observe electrocardiogram and oxidation of blood, while the patient is many kilometres away. In case of unstable state of health, this system is able to alert the physician. The teleconsultation system provides a videoconference service with shared multimedia document (audio-video streams, graphic, text, etc.).

7

Different approaches applied to the contextual medical decision problem

Kurzyński M., Puchała E., Woźniak M.

Journal of Medical Informatics & Technologies

|

2001

|

Vol. 2, Part. 1

MI147--MI153

EN

In many diagnostic problems there exist dependencies between successive states object and applied control. This situation is typical for the medical decision task, i.e. the recognition of the human acid-base state. We will present neural networks, probabilistic and fuzzy approach applied to the medical decision problem with context.

8

Recognition of fetus defects on the base of computer analysis of fetal heart rate signal

Kurzyński M., Morawiec B., Tomaszek M.

Journal of Medical Informatics & Technologies

|

2000

|

Vol. 1

BI 51--57

EN

The present paper deals with the experimental comparative analysis of method and algorithms of fetal heart rate (FHR) signal processing. The purpose of FHR signal processing is computer-aided recognition of uncomplicated pregnancy, pregnancy with arterial hypertension and with immunological conflict. In the discussed methods we can distinguish two steps: reduction of information and decision making. In the first step two different methods of time and frequency analysis of FHR signal were applied: Schur algorithm of linear optimal mean square prediction and morphological features describing acceleration and deceleration phenomena in the FHR signal. In the second step however, we used three-layer artificial neural network of back propagation type with sigmoid transfer functions and pattern recognition approach with minimum-distance algorithms. In the former method the following procedures were applied: nearest mean algorithm, nearest neighbour and 3-nearest neighbour algorithms. Experimental investigations using MATLAB 5.2 program were made and their results are presented in Table 1. These investigations which are sequel to the previous experimental studies proves that information contained in the FHR signal can be source of valuable and interesting knowledge on the fetus defects.

PL

Praca przedstawia wyniki eksperymentalnej analizy porównawczej wybranych metod i algorytmów przetwarzania sygnału częstości akcji serca płodu (FHR) w celu komputerowego wspomagania rozpoznawania ciąży fizjologicznej, ciąży powikłanej konfliktem serologicznym oraz ciąży powikłanej nadciśnieniem tętniczym. Przedyskutowane metody różnią się procedurami redukcji cech oraz algorytmami decyzyjnymi. Do wydobycia parametrów opisujących sygnał FHR wykorzystano metody przetwarzania w dziedzinie czasu i częstotliwości, a jako algorytmy klasyfikacji zastosowano minimalno-odległościowe algorytmy rozpoznawania oraz sieć neuronową propagacji wstecznej.

9

Use of the computer program for histopathological image analysis in assessment of AgNORs expression

Piasecki T., Jeleń M., Kurzyński M., Zalesska-Kręcicka M., Jabłonka A., Kręcicki T.

Image Processing & Communications

|

1998

|

Vol. 4, no 1-2

81-86

EN

Selecting certain objects within a microscopic image, counting is amount and area gives important information for a diagnosis and prognosis of a disease. This paper presents a computer program called Komórki for histopathological image analysis consisting in counting number and area of AgNORs. The mean number of AgNORs and AgNORs area estimated by program Komórki was comparable with results of other authors. The consistency of the results obtained by conventional method and computer program was also high. These facts prove usefulness of this program in clinical practice.