Suzana Loskovska - Academia.edu (original) (raw)

Papers by Suzana Loskovska

MRI classification is a very important field of research due to the area of its implementation. T... more MRI classification is a very important field of research due to the area of its implementation. The aim of this article is to compare support vector machines (SVM), k-nearest neighbors and C4.5 classifiers when they are applied to MRIs. The dataset used for classification contains magnetic resonance images classified in nine classes. All images of the dataset are described with seven descriptors. The analysis of the classifiers was made for each descriptor separately. According to experimental results we conclude that support vector machines are the most precise and appropriate for the MRI dataset used in this research.

The main goal of the paper is to explore hierarchical classification. The investigation is perfor... more The main goal of the paper is to explore hierarchical classification. The investigation is performed on the dataset of Magnetic Resonance Images (MRI) which is hierarchically organized. Generalized top-down hierarchical classification architecture is proposed in the paper. Additionally, two specific cases of the generalized architecture are explored: three-stage hierarchical architecture based on SVM and three-stage hierarchical architecture based on ANN. From the performed experiments, it is concluded that the SVM based scheme outperforms the ANN based scheme. Moreover, the gain of the investigation conducted in this paper becomes bigger with the possibilities given by the proposed generalized architecture for further investigations.

The aim of the paper is to compare classification error of the classifiers applied to magnetic re... more The aim of the paper is to compare classification error of the classifiers applied to magnetic resonance images for each descriptor used for feature extraction. We compared several Support Vector Machine (SVM) techniques, neural networks and k nearest neighbor classifier for classification of Magnetic Resonance Images (MRIs). Different descriptors are applied to provide feature extraction from the images. The dataset used for classification contains magnetic resonance images classified in 9 classes.

In the past few years, immense improvement was obtained in the field of content-based image retri... more In the past few years, immense improvement was obtained in the field of content-based image retrieval (CBIR). Nevertheless, existing systems still fail when applied to medical image databases. Simple feature-extraction algorithms that operate on the entire image for characterization of color, texture, or shape cannot be related to the descriptive semantics of medical knowledge that is extracted from images by human experts.

In order to help and support learning practices, the development of web-based Personal Learning E... more In order to help and support learning practices, the development of web-based Personal Learning Environments (PLE) is widely adopted. A PLE is a set of services customized by the student. Among these services, resource (either digital or human) recommendation is a crucial one. The paper briefly reviews existing approaches for recommending resources in PLE. Then it describes a novel approach implemented in the OP4L prototype. OP4L is combining Social Web presence data and semantic web technologies, based on an intensive use of ontological models to represent the learning context. Then the paper reports about qualitative studies that were conducted with students on the currently available version of the OP4L prototype. The aim of the study was to get students' feedbacks about new online presence services offered in a LMS.

Sarcoma

² The main goal of this paper is to propose and investigate a three-stage hierarchical classifica... more ² The main goal of this paper is to propose and investigate a three-stage hierarchical classification scheme. The proposed scheme was examined on the dataset of magnetic resonance images. The experimental results obtained from the classification with the proposed method outperformed the results provided by the flat classification and the two-stage hierarchical classification examined in our previous work. According to this, we can conclude that the proposed method is more suitable for solving MRI classification problem ...

Despite of the presence of many commercial OCR programs on the market, the recognition accuracy i... more Despite of the presence of many commercial OCR programs on the market, the recognition accuracy is still a difficult problem, especially when handling multilingual documents. In absence of an OCR program which can concurrently recognize mixed script documents, we have developed an OCR program for recognizing mainly Macedonian text with words in Latin script. In this paper an adaptation of preclassification based on character shape, for Macedonian characters along with standard Latin characters is presented. Preclassification is performed using the line parameters such as baseline and upper-baseline, as well as the dimensions of the character boxes and their horizontal position within the text line.

… Interfaces (ITI), 2010 …, Jun 21, 2010

… (ITI), Proceedings of …, Jun 27, 2011

Among the applications of computer science in the field of medicine, the processing of medical im... more Among the applications of computer science in the field of medicine, the processing of medical image data is playing an increasingly important role. With medical imaging techniques such as X-Ray, computer tomography, magnetic resonance imaging, and ultrasound, the amount of digital images that are produced in hospitals is increasing incredibly fast. Thus the need for systems that can provide efficient retrieval of images of particular interest is becoming very high. The two different approaches used for the representation of images are: the metadata-based and the content-based approaches. For very large database personally describing and annotating every image with text indices is time-consuming and impractical. In this paper we propose a content-based image retrieval system with parallel retrieval engines to achieve higher retrieval performance and efficiency.

Sarcoma

The objective of the paper is to explore classification on magnetic resonance images (MRI). In ou... more The objective of the paper is to explore classification on magnetic resonance images (MRI). In our work on MRI classification, two types of classification (flat and hierarchical) are addressed and explored. The examination is conducted on the dataset of magnetic resonance images that have hierarchical organization. All images are described by using Edge histogram descriptor for the feature extraction process. We compared the experimental results obtained from the hierarchical classification to the results provided by ...

ITI 2008 - 30th International Conference on Information Technology Interfaces, 2008

In this paper, we present the advantages from the use of data visualization from information syst... more In this paper, we present the advantages from the use of data visualization from information systems of communal arrangement department for the local selfmanagement. This approach will provide help to the users to analyze and see information more quickly and more efficiently. These visual representations are expected to shorten required time for data analyses. In the paper, we present several visualizations of the integrated data for managers and analytical staff in the local municipality in a city in Republic of Macedonia where the one-stop-shop system is still not implemented.

2009 2nd International Congress on Image and Signal Processing, 2009

... Ivica Dimitrovski, Suzana Loskovska Department of Computer Science and Computer Engineering F... more ... Ivica Dimitrovski, Suzana Loskovska Department of Computer Science and Computer Engineering Faculty of Electrical Engineering and Information Technologies Skopje, Macedonia ... 2. The system includes two main sub-systems: the server and the client sub-system. ...

MELECON '98. 9th Mediterranean Electrotechnical Conference. Proceedings (Cat. No.98CH36056), 1998

A model for implementing recursive objects, defined by L-systems, in a ray tracing based system f... more A model for implementing recursive objects, defined by L-systems, in a ray tracing based system for realistic visualization is presented. This model is based on constructive solid geometry (CSG). Recurrent CSG-graphs are used for internal representation of recursive objects. These graphs allow one to build up the scene during visualization, i.e. generating only those primitive objects which might be affected

A subsystem for text-to-speech (TTS) conversion for Macedonian language as a part of a system for... more A subsystem for text-to-speech (TTS) conversion for Macedonian language as a part of a system for support of humans with damaged sight will be presented in this paper. The whole system includes recognition of printed Cyrillic text, archiving, TTS conversion and printing on a Braille printer. None of tested commercial programs for generating human speech on personal computer is appropriate for Macedonian language. Therefore a subsystem for real-time TTS conversion from unrestricted text is under development. The whole architecture of the subsystem and some of it's major modules will be presented here. Number of techniques, like automatic generation of prosodic rules, time scale modification of speech based on nonlinear oscillator model and flexible speech units inventory will be used to overcome the fact that Macedonian language is comparatively poorly researched

Shot boundary detection is fundamental to video analysis since it segments a video into its basic... more Shot boundary detection is fundamental to video analysis since it segments a video into its basic components. This paper presents a comparison of several shot boundary detection techniques and their variations including color histogram, edge directions histogram and wavelet transformations statistics. The performance and ease of selecting good thresholds for these algorithms are evaluated based on a wide variety of video sequences with different object and camera motions. Threshold selection is performed using sliding window. We used TV news, sports and documentary, music, movie and nature video sequences to estimate the performance of the algorithms. The experimental results indicate that the algorithm based on color histograms is most suitable for shot boundary detection in film and documentary categories, but the algorithm based on wavelet is preferable for nature and sports categories.

Sarcoma

… Interfaces (ITI), 2010 …, Jun 21, 2010

… (ITI), Proceedings of …, Jun 27, 2011

Sarcoma

ITI 2008 - 30th International Conference on Information Technology Interfaces, 2008

2009 2nd International Congress on Image and Signal Processing, 2009

MELECON '98. 9th Mediterranean Electrotechnical Conference. Proceedings (Cat. No.98CH36056), 1998