Marzena Bielecka

Wind turbines states classification by a fuzzy-ART neural network with a stereographic projection as a signal normalization

In this paper wind turbines operational states classification is considered. The fuzzy-ART neural network is proposed as a classifying system. Applying of stereographic projection as an input signals normalization procedure is introduced. Both theoretical justification is discussed and results of experiments are presented. It turns out that the introduced normalization procedure improves classification results.

Application of shape description methodology to hand radiographs interpretation

In this paper, a shape description methodology, introduced by Jakubowski [6] is applied to hand radiographs interpretation, in order to recognize bones borders shapes in the fingers regions. It is shown that the classical approach can be used only for preliminary analysis. Therefore, the improved method, based on fuzzy approach, is considered.

ART-2 Artificial Neural Networks Applications for Classification of Vibration Signals and Operational States of Wind Turbines for Intelligent Monitoring

In recent years wind energy is the fastest growing branch of the power generation industry. The largest cost for the wind turbine is its maintenance. A common technique to decrease this cost is a remote monitoring based on vibration analysis. Growing number of monitored turbines requires an automated way of support for diagnostic experts. As full fault detection and identification is still a very challenging task, it is necessary to prepare an “early warning” tool, which would focus the attention on cases which are potentially dangerous.

Modified jakubowski shape transducer for detecting osteophytes and erosions in finger joints

In this paper, a syntactic method of pattern recognition is applied to hand radiographs interpretation, in order to recognize erosions and osteophytes in the finger joints. It is shown that, the classical Jakubowski transducer does not distinguish contours of healthy bones from contours of affected bones. Therefore, the modifications of the transducer are introduced. It is demonstrated, that the modified transducer correctly recognizes the classes of bone shapes obtained based on the medical classification: healthy bone class, erosion bone class and osteophyte bone class.

A fuzzy shape descriptor and inference by fuzzy relaxation with application to description of bones contours at hand radiographs

Generalization of string languages describing shapes in order to apply them to analyze a contour of bones in hand radiographs is proposed in this paper. An algorithm to construct a fuzzy shape descriptor is introduced. Next, basing on the fuzzy descriptor, a univocal description by fuzzy interference is realized. In prospects this method will be used to erosion detection of hand bones visible in hand radiographs.

Generalized Shape Language Application to Detection of a Specific Type of Bone Erosion in X-ray Images

X-ray imaging is crucial in diagnosis of various musculoskeletal diseases. During early disease process, the X-ray changes are often scarce and difficult to capture and the definite localization of osteophytes or erosions is often challenging. Therefore, the attempt to use computer methods to facilitate better diagnosing is of great value. Formal tools for contour description are based on string languages. In Jakubowski’s shape languages sixteen primitives are predefined. Finite collection of primitives, however is insufficient for describing natural objects because of irregular character of this type of objects. In this paper the generalized shape language, in which primitives are defined on a higher level of abstraction, is proposed and is used for description and detection of a special type of complex erosions in bone contours.

Hybrid System of ART and RBF Neural Networks for Classification of Vibration Signals and Operational States of Wind Turbines

In recent years wind energy has been the fastest growing branch of the power generation industry. Maintenance of the wind turbine generates its the largest cost. A remote monitoring is a common method to reduce this cost. Growing number of monitored turbines requires an automatized way of support for diagnostic experts. Early fault detection and identification is still a very challenging task. A tool, which can alert an engineer about potentially dangerous cases, is required to work in real-time. The goal of this paper is to show an efficient system to online classification of operational states of the wind turbines and to detecting their early fault cases. The proposed system was designed as a hybrid of ART-2 and RBF networks. It had been proved before that the ART-type ANNs can successfully recognize operational states of a wind turbine during the diagnostic process. There are some difficulties, however, when classification is done in real-time. The disadvantages of using a classic ART-2 network are pointed and it is explained why the RBF unit of the hybrid system is needed to have a proper classification of turbine operational states.

Improved fuzzy entropy algorithm for x-ray pictures preprocessing

The fuzzy entropy algorithm was designed for preprocessing of photos taken in the visible spectrum of light. However it did not produce satisfying results when it is directly applied to X-ray pictures. In this paper we present significant improvements of this approach and apply it to hand radiographs. The noise elimination and the bone contourisation is the task which is studied in this paper. Not only is the algorithm modified but also it is combined with using of median and minimum filters. The presented approach allows us to obtain satisfying noise elimination and clear bone contourisation.

Optimization of Numerical Calculations of Geometric Features of a Curve Describing Preprocessed X-Ray Images of Bones as a Starting Point for Syntactic Analysis of Finger Bone Contours

Analysis of bone contours in X-ray images is crucial for the detection of pathological changes such as erosions and osteophytes. The analysis is done by using shape languages. In this approach the contour received from the preprocessing procedure is segmented into fragments according to geometrical properties of the contour. The properties are characterized by monotonicity and convexity of the contour. Two aforementioned features are deduced by using the first and second derivatives that are calculated numerically. On the one hand the used numerical procedure can smooth the analyzed contour. On the other hand, however, the more smoothed the contour is, the more chance that the small pathological changes remain undetected. Finding the optimal numerical procedure for X-ray hand images is the aim of this paper. (This paper was supported by the AGH - University of Science and Technology, Faculty of Geology, Geophysics and Environmental Protection as a part of the statutory project).

Automatized fuzzy evaluation of CT scan heart slices for creating 3D/4D heart model

Graphical abstractDisplay Omitted HighlightsA new method of evaluation of CT scan heart slices is proposed.The proposed method is based on classification of histograms of brightness of CT scan heart slices.The quality evaluation is based on fuzzy classification.The algorithmic approach to the construction of the membership functions of given in advance classes is introduced.The experiment showed that the fuzzy selection is absolutely consistent with the one done by an expert. A new method of evaluation of CT scan heart slices is proposed in this paper. CT images, acquired from different patients, are stored in PACS database. In order to create 3D/4D model of heart it is necessary to choose these CT images that have sufficient quality. The proposed method is based on classification of histograms of brightness of CT scan heart slices. Some structural features of these histograms are correlated to the images quality which is evaluated in the context of creating an ultrasonography simulator on the basis of CT scan heart slices. They constitute computed tomography scan sets. The quality evaluation is based on fuzzy classification. A new methodology of the membership function construction in relation to structural features of the examined images is proposed. The algorithmic approach to the construction of the membership functions of given in advance classes is introduced. The experiments have shown that the proposed method is effective in selection of high quality CT scan heart slices that can be the basis for the simulator construction. The experiment showed that the proposed fuzzy selection is absolutely consistent with the one done by an expert.