Abstract: In this study different methods of medical imaging are considered along with the possibility of combining information offered by them. One of the techniques of special interest is positron emission tomography (PET). This medical imaging technique gives information about metabolic processes in the human body. One of the main defiance of this method is attenuation of gamma photon in interaction with tissues and organs inside the body. This phenomenon can be fixed by the attenuation correction factor. The factor can be obtained using Computerized tomography (CT) imaging with aim to map PET image with linear attenuation coefficient. The PET and CT images are matched using image registration technique which provides determination of attenuation coefficient space distribution. Using the attenuation coefficient the pixels intensity of PET image can be scaled by the reciprocal value of attenuation coefficient with propose of attenuation correction.

Abstract: This paper presents a research that is developed inside “Products and Systems Design and Development” research group of Pontifical Catholic University of Parana (PUC-PR). In this research is studied, designed and developed a DICOM image loader with agents based system, that in a near future will be a important module of a dental prosthesis design decisions multi-agents software. In this paper, will be shown an overview of agent systems explaining its structure with a framework called JADE, used to provide agents software environment. The methods used to guide this research were, the bibliographical survey to list de concepts, and, states of art for the expert system project layout, develop and tests procedures. As a result, this paper shows that using agents’ software techniques in expert systems development is a new way to manage data in a flexible way, offering forms to creating modular solutions that should have the capacity to deal with complex scenarios. It also allows, better software scalability and modularization for complex and specialized software solutions.

Abstract: Sharing different types of models over a long distance using Internet has great potential as a resource for research and education. This paper presents the development of different 3D model types of human spine and web-based application for user access and interactive exploration. The remote visualization of models is powered by O3D plug-in/WebGL, GLview 3D Plugin, and STLDroid. End-user is able to review online in real time, and download high-definition 3D models of spinal vertebrae. The presented solution could improve education in the field of biomedical engineering and become useful tool in spinal research.

Abstract: Fuzzy XQuery is the extension of standard XQuery language that allows fuzzy values in the query condition statements. Relational operators are not only required and possible in crisp value cases but also for fuzzy values. When relational operators are included in the query, it is necessary to provide means for comparison between fuzzy sets. These fuzzy relational operators are typically used in two fuzzy sets comparison case, but can also be used with some aggregate functions like MIN, MAX, and SUM. The aim of this paper is to present the algorithms for the implementation of fuzzy relational operators. Our algorithms compare the horizontal positions of two fuzzy sets and calculate the ordering value based on partial fuzzy ordering proposed by Bodenhofer. Moreover, we developed a GUI application and evaluated our approach with 360 fuzzy ordering cases. The experimental results show that our algorithms are capable of calculating fuzzy ordering values with various types of fuzzy values correctly.

Abstract: R is considered both as a programming language and an environment for statistical computing. It provides a wide variety of functionalities and almost limitless opportunities to extend. Its easiness to use, in combination with the interpreted nature, ensures its popularity in different areas of business and science. However, when it comes to more demanding computations in terms of performance, R turns to be significantly slower, compared to other languages. As the need for data analysis increases rapidly, the existence of a simple and at the same time efficient tool is of inestimable importance. This paper strives to present an analysis of performance drawbacks of R, and exposes an idea for possible improvements.

Abstract: Agile approach to software development is formally established in 2001 with Agile Manifesto promotion. Many different methods of agile approach were used by practitioners for many years and the need for their integration and tailoring has emerged to new methodology created by IBM and promoted since 2012 as disciplined agile delivery approach. This paper aim is to describe the role of modeling in disciplined agile approach. A case study is presented with results of project of information system development within educational environment. Selection of models to be used for information system development within disciplined agile approach in this case study is oriented towards business process model, UML's use case model and data models. They are represented as a basis for agile software development and iterative delivery within the case study. CASE study is focused on the analysis of the impact of business process modeling to information systems software project performance within Disciplined Agile Delivery approach.

Abstract: The agent technology represents one of the most consistent approaches to the development of distributed systems. Multiagent middleware XJAF, developed at the University of Novi Sad, presents a runtime environment that supports the execution of software agents. To solve the problem of interoperability, we propose a domain-specific agent language named ALAS, whose main purpose is to support the implementation and execution of agents on heterogenous platforms. To define the structure of the language, a metamodel and a grammar of the ALAS language has been created, in accordance with the requirements and needs of the agents. This paper describes the construction of the compiler and the generation of executable Java code that can be executed in XJAF.

Abstract: The paper presents an overview of techniques and mechanisms implemented in generic web and desktop engines that enable execution of application prototypes being specified by our Kroki tool. Unlike most other solutions where only a user interface skeleton is executable, Kroki’s specifications can be tested through all three application tiers – the user interface, the business logic, and the database. Kroki is a mockupdriven tool that enables development of enterprise information systems based on participatory design. Since immediate execution is always possible, it can significantly contribute to decreasing a communication gap between the development team and users.

Abstract: In order to achieve modern business requests such as shorten time of advancement and launching the new products and services as well as end users inclusion in product development process, specific methodologies need to be applied. This paper presents application of the Scrum methodology in software project realization. It is the real world Telenor Serbia E-business software project described from every aspect of the Scrum implementation beginning with the project start, then through pilot project, sprint execution, usability testing and concluding remarks. This may be used as an example in other businesses and companies in order to improve project realization and business operations.

Abstract: In this paper java agent architecture for utilizing available multi-core and many-core hardware is presented. Architecture is developed using JADE (Java Agent DEvelopment Framework) and OpenCL standard for programming multi-core and many-core devices. A JADEbased system can be distributed across available machines and OpenCL promises portability of the developed code between heterogeneous devices. OpenCL is an open, royalty free, standard developed by Khronos group for parallel programming of heterogeneous devices (CPU’s, GPU’s, … ) from different vendors. Developed java agents communicate with each other in accordance with FIPA specification and are independent from operating system of the specific machine they run on. In this paper two java agents are presented. First agent inspects available hardware with OpenCL support and sends message to the second agent which runs simulation utilizing all the available devices with OpenCL support discovered by the first software agent.

Abstract: Integration of dissimilar and heterogeneous data sources is a continuous challenge for life science investigation. For the researchers in a domain of life science, finding the relevant information across different data sources is of crucial importance. Current researches strongly depend on availability, accessibility and effective data use. Semantic Web technologies enable facilitated collected data aggregating. Extracting data from large, semantically based datasets became a trying challenge for a researcher who needs to put an effort in discovering each connection in the datasets of interest, different properties among datasets and classes, name conventions, as well as to understand precise meaning of every enumerated part. For systems without specific conventions in organizing separate datasets, these tasks can be quite complex and timeconsuming. In this paper, we present the challenges of data integration of same-domain datasets and the use cases that identify our approach aimed at finding the relevant data essential for our further research. We have developed a semantically based web application that utilizes the ontologies and SPARQL queries for the data process search and integration in molecular biology research center. For selected substance from our semantic dataset we have generated SPARQL queries for data discovering by means of the existing and constructed templates. In this way a knowledge database for valid and tested SPARQL queries is created and presented in a new ontology, established for this purpose.

Abstract: In recent years people are increasingly opting for energy self-sufficiency of their residential and commercial buildings. The application of solar energy is one of the most attractive options, since the installation of photo-voltaic systems is fairly easily. The question is often on which parts of the roof is best to put such systems to achieve the most optimal use. New public e-services SolarEnergo is constructed and visually shows how much the roof is suitable for the installation of photo-voltaic systems and gives detailed calculation on the amount of solar energy throughout the year. The calculation takes into account also shading from other buildings, terrain within a radius of 15 km and high vegetation. The project has successfully acquired funds from the Ministry of Education, Science and Education in Slovenia and European Regional Development Fund for innovative IT solutions. For the development Municipality Beltinci is cooperating with the Faculty of Electrical Engineering and Computer Science, where the experts have the necessary knowledge and experience to develop advanced solutions in the field of geographic information systems (GIS). The accuracy of topology information and information about high vegetation is achieved using data from airborne laser scan technology LiDAR (Light Detection and Ranging), which scans terrain and buildings at a resolution of 20 points/m2. For the development of the web and mobile e-service open source platforms in the field of GIS are used. Thus, we have shown that even with less expensive technology, reliable and advanced applications can be developed. The result is a new web and mobile e-service SolarEnergo that allows citizens to easily select the object also with the help of smart Android phones and GPS positioning. With this, Municipality Beltinci offers a new e-service, which can provide, directly in the field or from the office, all the necessary information to decide on the use of solar energy. In the next step SolarEnergo can be easily upgraded and extended to other municipalities or also to whole countries.

Abstract: Supplier search can be daunting for manufacturers, given the vast number of suppliers available and the work required to filter through all the information. However, even if this process allows the selection of a few potential suppliers of interest, dealing with potential suppliers for the first time may bring uneasiness, as there is no guarantee that they will comply with the requested products or services. On the other hand, suppliers which want to be seen and differentiate themselves in order to win more customers also face similar challenges. The vast amount of competitors can make it extremely difficult for any given supplier to stand out from amongst its competitors. Also, new entrants into the market with little or no reputation may find it difficult to attract their first customers. The IMAGINE project is developing a methodology and a software platform for the management of the dynamic manufacturing networks which provides innovative Supplier Search capabilities. The present paper provides an overview of the developments behind the IMAGINE platform and proposes a framework in order to help the challenges that suppliers and manufacturers face.

Abstract: In this paper we propose a solution for the development of enterprise portal-like web applications that is based on the Liferay portal and the Alfresco document repository. Combining a robust and featurerich platform for building the application layer such as Liferay with the industry standard document management system such as Alfresco enables developers to rapidly provide their customers with a custom portal web application that suit their needs. Emphasis of this paper is on extending the combined solution by implementing a custom activity list that shows integrated user activities from both Liferay and Alfresco, a feature missing in the original combined platform.

Abstract: Over the recent years, international public experiences increased occurrence of natural and man-made disasters involving large human casualties and fatalities, displacing affected population, and bringing immense material loss. To cope with such devastating and highly unpredictable events, disaster managers and first responders participating in the crisis aftermath, need all the existing technological help to be able to provide timely and accurate response, better coordination, and overall risk prevention and mitigation. In this paper we provide an overview of current state of the art commercial, opensource, and research emergency management tools aiming for recovery phase: disaster in-situ data collection, communication and DSS (decision support system) tools. An overview of general requirements for such critical tools is given as well and some integration challenges identified.

Abstract: In today’s volatile and turbulent business environment, supply chains face great challenges when making supply and demand decisions. Making optimal inventory replenishment decision became critical for successful supply chain management. Existing traditional inventory management approaches showed as inadequate for these tasks. Current business environment requires new methods that incorporate more intelligent technologies and tools capable to make accurate and reliable predictions. This paper deals with data mining applications for the supply chain inventory management. It describes the use of business intelligence (BI) tools, coupled with a data warehouse to employ data mining technology to provide accurate and up-to-date information for better inventory management decisions and to deliver this information to relevant decision makers in a user-friendly manner. Experiments carried out with the real data set showed very good accuracy of the model which makes it suitable for more informed inventory decision making.

Abstract: To remain competitive and agile, modern organizations invest several business and IT enablers in order to supervise their change management initiatives. In this perspective, research in enhancing business process management (BPM) capabilities (maturity, risk assessment, etc.) presents relevant guidelines in order to adapt IT solutions when business requirements evolve. In this context, this research proposes an ontology based framework for business processes assessment. The first objective of this research covers the definition of a functional category of ontological concepts related to the qualification of business process dimensions. Hence, we identify concepts related to processes lifecycle, process maturity, task related risk, etc. The second category covers the non-functional aspects related to the QoS, event management, etc. The second objective targets the definition of assessment model based on the instantiation of the proposed ontology. The application of this approach aims to identify the stark nodes of business processes and to validate the alignment with evolutionary business processes requirements.

Abstract: Hydro information systems in Serbia are in constant development from the early 80’s. Considerable advancements have been made in application of modern methodologies and technologies, with practical use in hydro power and water resources management. In order to provide support to management of complex hydropower systems and their digital simulations it is necessary to establish communication between measurement systems and computational models, although they were not primarily designed on the concepts of mutual integration and possible reusability. This paper presents a solution for distributed hydro-meteorological and hydropower data acquisition, simulation and design support that is based on serviceoriented architecture (SOA) principles. The complex and heterogeneous information flow used by these systems is portioned into separate data and computational functions that use common open protocols for mutual communication. Based on these techniques, the paper presents the design of SOA hydro-information system and its application in case of Drina river basin management.

Abstract: Application of demand-side management (DSM) plays today an important role in energy management both in industrial and residential domain. This paper proposes a generic approach to DSM which is based on the genetic algorithm (GA), as one of the powerful search heuristics inspired by the process of natural evolution. The proposed approach was defined flexibly enough to be capable of discovering an optimal load distribution (e.g. from the financial perspective) in practically any multiple energy supply/multiple load facility infrastructure. Optimization of the demand side was carried out by taking into account the forecasted energy demand and applied tariff schemes as well. Furthermore, a performance of the proposed approach was verified at a multiple supply/multiple load use case scenario. Based on the optimization results, it was concluded that the proposed GA based solution could be successfully utilized to facilitate decision making of energy managers regarding the appropriate DSM measure selection.

Abstract: Permanent increase of energy prices united with greater energy demand make the reduction of energy infrastructure operation costs a challenging task. Therefore, systematic application of energy management (EM) solutions became a necessity and the most viable approach for cutting down the energy costs. This paper proposes an EM solution that brings advancement of the state of the art related to the multi-carrier energy dispatch by optimizing the energy flows within a generic energy infrastructure. The proposed advancement is achieved through the enrichment of existing Energy Hub concept, leveraging on supply side optimization, with complementary optimization of demand side, known also as demand side management (DSM). Compelling simulation results, justifying the merging of these two concepts, were reported. The results show the potential for saving up to 25% of energy costs, depending on the use case scenario, comparing to a baseline scenario where no EM solution is applied. Considering that the proposed solution employs existing energy infrastructure, thus does not require expensive equipment retrofit, it makes it immediately applicable in both residential and commercial domain.