Luka Pavlič

Medical diagnostic process optimization through the semantic integration of data resources

In this paper we study the optimization of medical diagnostic process from the data access point of view. According to many studies which showed that optimized diagnostic process can considerably improve efficiency in health care industry, we present a new approach to data integration within a diagnostic process. It is our belief that a unified access to data resources throughout the whole diagnostic process considerably improves the efficiency of the process itself. When combining the optimized data access with an existing algorithmic optimization method an optimized process can be achieved that takes into account the quality of a diagnosis, the individual needs of each patient, the associated costs, and the utilization of personnel/equipment. To enable an efficient management of data, we developed a semantic web based system for the integration of data resources within a medical diagnostic process. Then we combined the unified data access with our existing diagnostic process optimization framework that uses machine learning techniques and evolutionary algorithms. The new defined diagnostic process framework is finally used in a case-study for optimizing the diagnosing of the mitral valve prolapse syndrome in a regional hospital department.

Improving design pattern adoption with Ontology-Based Design Pattern Repository

Design patterns are a proven way to build high-quality software. The number of design patterns is rising rapidly, while management and searching facilities seems not to catch up. This is why selecting a suitable design pattern is not always an easy task. This issue is especially clear for less experienced developers. In this paper we present our approach to cope with the presented issue - an experiment prototype of a new design pattern repository, based on semantic web technologies. Since new ontology-based design pattern repository is a work in progress we point out its potentials for improving design pattern adoption.

A question-based design pattern advisement approach

Design patterns are a proven way to build flexible software architectures. But the selection of an appropriate design pattern is a difficult task in practice, particularly for less experienced developers. In this paper, a question based design pattern advisement approach will be proposed. This approach primarily assists developers in identifying and selecting the most suitable design pattern for a given problem. We will also propose certain extensions to the existing Object-Oriented Design Ontology (ODOL). In addition to the advisement procedure, a new design pattern advisement ontology will be defined. We have also developed a tool that supports the proposed ontology and question-based advisement (OQBA) approach. The conducted controlled experiment and two surveys have shown that the proposed approach is beneficial to all software developers, especially to those who have less experience with design patterns.

Supporting Collaboration of Medical Informatics Researchers and Teams

Medical informatics is a highly interdisciplinary field of research, where a great deal of collaboration is needed to achieve efficient solutions. We are witnessing a remarkable proliferation of biomedical knowledge on one side and the revolution of communication technologies on the other. In this manner we may expect that a proper methodology to support collaboration strategies of medical informatics researchers and research teams based on knowledge, should improve the collaboration and the creativity in the field of medical informatics research. Based on the required functionalities of such a system, we propose the semantic web as the underlying technology and explain its use for knowledge sharing both within an organization and inter-organizationally.

Digital Production Pipeline for Virtual Cultural Heritage Applications Using Interactive Storytelling

Preserving the past within the collective memories of people is one of the ultimate goals of creative industries. Virtual cultural heritage applications recreate history by inviting users to travel back in time. One of the more popular approaches for enhancing the immersion of time-travellers within virtual environments is interactive digital story-telling, which enables users to learn while exploring. While the process of developing interactive digital story-telling applications is still complex, involving professionals from various artistic and scientific disciplines, significant technological advances have recently been made towards improving content production. Namely, advanced Earth observation systems are capable of capturing hundreds of thousands of points within a second, thus creating extremely accurate representations of artefacts such as cultural sites. Usually mounted on mobile and airborne platforms, they produce high-resolution point-clouds with densities ranging up to several hundred points per square meter. However, the lack of topology, huge data sizes, and the contained noise, requires new data processing, storage, and management approaches before these data can be successfully utilized. This chapter introduces an advanced pipeline for creating virtual worlds as environments for interactive digital story-telling applications. By considering the challenges and issues arising during this process, recent developments in content creation, warehousing, editing, and postproduction will be demonstrated through the usage scenario of the implemented end-user application.

Agile Coaching: The Knowledge Management Perspective

Currently, almost half of IT projects are managed in a traditional, sequential, plan-based manner. The leading development methods, successful for decades, are based on the so-called waterfall model. However, rapidly changing business environments push software development companies to adapt newer, leaner and more agile development methods. They promise development teams to welcome changing requirements, deliver software quickly and respond to new customer requests instantly. This is why transition to agile, mostly scrum-based development methods have been common lately. In order to avoid learning from their mistakes, companies usually seek help while adopting agile practices, related knowledge and experiences. Professionals, Agile Coaches, offer it.

An Entropy-Based Algorithm for Proposing a Suitable Design Pattern

This paper deals with the problem of selecting a suitable design pattern when necessary. The number of design patterns has been rapidly rising, but management and searching facilities appear to be lagging behind. In this paper we will present a platform, which is used to search for suitable design patterns and for design patterns knowledge exchange. We are introducing a novel design pattern proposing approach: the developer no longer searches for an appropriate design pattern, but rather the intelligent component asks the developer questions. We do not want to invest extra effort in terms of maintaining a special expert system. Guided dialogues consist of independent questions from different sources and authors that are automatically combined. The enabling algorithm and formulas are discussed in detail. This paper also presents our comparison with human-created expert systems via a decision tree. Experiments were executed in order to verify our approach performance. The control group used a human-created expert system, while others were given a proposing component to find appropriate design patterns.

LiDAR based terrains for virtual cultural heritage applications

Virtual presentation of cultural heritage has shown a great potential in preservation of collective memory. The level of immersion that the visitor feels in a virtual environment is one of the key factors for the quality of his experience. Terrains are the most authentic elements of cultural heritage sites, as they rarely have been significantly changed through time. In this paper we explore the use of LiDAR based terrains in interactive virtual cultural heritage applications, considering the advantages and drawbacks of this approach. Our case study is the virtual presentation of the Early Christian Basilica in Cim near Mostar, created for the interactive exhibition “Keys to Rome”.

Entropy- and Ontology-Based E-Services Proposing Approach

E-services have significantly changed the way of doing business in recent years. We can, however, observe poor use of these services. There is a large gap between supply and actual e-services usage. This is why we started a project to provide an environment that will encourage the use of e-services. In the paper we propose an idea of intelligent e-services platform. In addition to established possibilities of searching (e.g. keyword searching, manual classified knowledge browsing), we also propose our own original approach. The ontology and algorithm for proposing appropriate e-services are described in the paper. We use expert knowledge in form of question-answer pairs. It is used by the algorithm to dynamically guide a dialog with user. Intelligently selected sequence of questions is used to suggest the e-service that could help user at a given situation. Ontologies and semantic web technologies are used heavily therefore.

Metadata for research and development collaboration between universities

Knowledge Management (KM) research requires expertise from many different disciplines. To help with KM research activities of different groups that are geographically distributed, a new Knowledge Management in Organisations (KMO) structure is proposed. The organisational structure of KMO is based on decentralised units where autonomy and independence are established in each group. Effective design of the system requires that a proper format for representing and organising the shared knowledge is crucial. This paper describes the development of metadata and ontology for our KMO system to support the sharing of knowledge for the distributed partners between the four universities.