Kostas Vassilakis

Decision Support System for Classification of Epilepsies in Childhood

Diagnosis of epilepsy in childhood is often difficult as the symptoms are often atypical and the epilepsy syndromes are multiform. Methods from the domain of artificial intelligence give the opportunity to formalize medical knowledge and standardize various diagnostic procedures in specific domains of medicine. We developed a decision support system using artificial intelligence techniques for the classification and ultimately the diagnosis of epilepsies and epilepsy syndromes in children. The system incorporates knowledge from the International Classification of Epilepsies and Epileptic Syndromes. It was assessed using clinical data and the systems conclusions were compared with the diagnoses proposed by an experienced doctor. The system and the physician reached identical diagnoses in 85.2% of the cases. In an additional 8.2% of the cases, the systems diagnosis was similar to that of the physician, thus raising its overall success rate to 93.4%. The system can be helpful, especially for trainees, since it only needs to import the clinical and laboratory data. Decision making and differential diagnosis are then performed automatically. (J Child Neurol 2002;17:357-363).

An Educational Model for Asynchronous E-Learning. A Case Study in a Higher Technology Education

Nowadays, the use of Web based Education (WbE) in distance learning education is considered to be an innovative method of learning. Supportive parties argue that WbE renews the educational practice through the use of computers and their applied methodology, as well as the technologies provided by the use of the internet. These result in consciously renewing the educational material and in creating a flexible structure which promotes the individualization of learning. The proposed model aims at delivering technological classes through the internet, offering a flexible use of means and tools, allowing a synthetic presentation of selected bibliographic texts that cover the whole cognitional object, developing a cooperative spirit and individualising the learning procedure.

Extending the Sana Mobile Healthcare Platform with Features Providing ECG Analysis

The great development of technology recently provides innovations that improve everyday life. The major benefit of it is that medicine is also affected, so better healthcare can be provided. In that context, it can be critical for patients who suffer from chronic heart diseases to have in their availability a system that can monitor and analyse their electrocardiogram (ECG) displaying either normal or abnormal findings. The current chapter describes such a system that uploads, stores, processes and displays an ECG, calculating certain ECG findings necessary for doctors to make a diagnosis. To this end, the SANA mobile healthcare platform, with its OpenMRS open source enterprise electronic medical record system, has been chosen and extended in this work for storing, processing and displaying the ECG data. OpenMRS provides a user-friendly interface and a database for collecting medical big data. Analysis of ECG signals is leveraged by the Physionet toolkit. Physionet contains many ECG databases and the WFDB software for processing ECG signals. According to the scenario we have processed, an ECG is uploaded onto OpenMRS platform using a mobile device or any other Internet-enabled device and is stored in the database that OpenMRS uses. Then, ECG signal is filtered using a finite impulse response (FIR) filter to remove noise and using WFDB functions it is processed so certain intervals are determined. Finally, with the appropriate algorithms specific ECG findings are calculated. When the procedure completes, the results are stored into the database using SQL Queries. Using an HTML Form results and graphs are integrated into the OpenMRS website highlighting abnormal values with red color. Authorized users can have access to this information through any web browser.

Electronic health record with computerized decision support tools for the purposes of a pediatric cardiovascular heart disease screening program in Crete

BACKGROUND AND OBJECTIVE Early detection of cardiovascular (CV) disease or associated risk factors during childhood is of paramount importance, allowing for early treatment or lifestyle modifications, respectively. The objective of this study was to describe the development of an electronic health record (EHR), with integrated computerized decision support system (CDSS), specifically designed for supporting the needs of a pilot pediatric CV disease screening program applied on primary school students of a Mediterranean island. METHODS Evidence-based knowledge, national and international practice guidelines regarding sport preparticipation CV screening of children and young athletes has been used for the design of the designated EHR. A CDSS, capable for providing alerts for further cardiology evaluation need, has been incorporated into the EHR, based on normative anthropometric and electrocardiographic data as well as predefined positive history responses. RESULTS We developed a designated EHR with integrated CDSS supporting pediatric CV disease screening, capable for documenting CV-related personal and family history responses, physical evaluation data (weight, height, blood pressure), allowing for entering electrocardiogam (ECG) measurements and for uploading of multimedia files (including ECG images and digital phonocardiogram audio files). The EHR incorporates clinical calculators and referral alerts for the presence (and degree) of adiposity, hypertension, ECG abnormalities and positive history responses indicative of high CV disease risk. In a preliminary EHR validation, performed by entering data from 53 previously available paper-based health records, the EHR was proven to be fully functional. CONCLUSIONS The pediatric cardiology EHR with CDSS features which we developed might serve as a model for EHR for primary health care purposes, capable to document and early detect CV disease and associated risk factors in pediatric populations.

Implementing an Adaptive Learning System with the Use of Experience API

With the evolution of e-learning and its transformation into mobile learning, SCORM fails to keep up with learner’s need to discover knowledge through multiple and diverse sources. ADL’s Experience API (xAPI) fills this gap and offers a novel and flexible way to keep track of a learner’s activities and progress. In this paper, the xAPI and the concept behind it are shortly discussed, a brief comparison with SCORM is attempted and an innovative implementation of an adaptive LMS-free learning system with the use of xAPI is presented.

Real Time Evaluation of Education Methods via Smart Mobile Technology

We design, implement and evaluate performance of Exantas application which is compatible with Android Operating Systems Smartphone devices. As Exantas tool was able to show ancients travelers the correct route to follow we show that our application can help educational staff to improve their skills and evaluate on the fly how efficient is the educational style that they follow. Results can help teachers measure while teaching how much of the lessons content has been successfully absorbed by students and what are the topics that need further analysis or even a completely new explanation approach. As experiments show, Exantas is able to reduce teaching efforts and to reveal real lessons comprehension status since Teachers can make multiple Questions to all students and receive answers in seconds. Moreover, all answers are processed anonymously ensuring anonymity and integrity since students are not afraid to provide their actual answer.

Training the Mind: The GARDINER Platform

Recent research has shown that the systematic involvement of a person with games, which are designated to exercise memory and concentration, contributes to the long-term preservation of the human memory and therefore leads to the prevention of dementia. Our work seeks to capitalize on the positive effects of serious games’ use in a variety of ways. In particular, we provide insights into the design and development process of two serious games dedicated to being used by elderly people with dementia. In their context, we also elaborate on the basic elements of a novel web-oriented platform, namely GARDINER (Games plAtform foR minD traIning aNd mEmory peRk), aimed at making available various memory games which may have been crafted by various sources. Finally, some empirical data derived from the use of our platform and games in practice is provided.

Learning History Through Location-Based Games: The Fortification Gates of the Venetian Walls of the City of Heraklion

Games in education have always been a tool for increasing motivation and interest of learners. We present Location-Based Games (LBG) as a tool to involve and motivate students in the learning process. LBGs require the player to move around in order to complete a task and proceed in the storyline and use localization technology such as Global Positioning System (GPS). LBGs are built on physical worlds, while virtual world augmentations enable the interaction of physical and other related (cultural, historical etc.) data with the player. Augmented reality (AR) is used to provide this extra layer with 3D objects, avatars and animations for player’s interaction. In our paper we present a history learning LBG with the use of augmented reality in the form of 3D objects. We explore the concept, of having both virtual and physical worlds available within the same visual display environment.

Facilitating Learning in Isolated Places Through an Autonomous LMS

Current research argues that eLearning and mobile learning are forms of learning that could take place outside the classroom and the traditional learning environments. In addition, recent advancement in technology and increased use of smart devices permit students to carry with them a kind of portable smart device. Inevitably, sooner or later, these devices will become integral educational tools, such as pencils and books, while learning outside the classroom will continue to gain popularity as another form of learning. Ubiquitous learning aims to stimulate the wide use of ICT in Education and the enactment of autonomous digital resources for Outdoor learning. Technology could provide innovative ways of conducting outdoor courses, encompassing knowledge and physical activity. This paper presents the eClass-Pi system that facilitates outside the classroom eLearning and m-learning educational processes. It provides all the functionalities of a typical Learning Management Systems as well as synchronous and asynchronous teaching, portability and energy autonomy.

Ontological modeling of educational resources: a proposed implementation for Greek schools

In eLearning context searching for suitable educational material is still a challenging issue. During the last two decades, various digital repositories, such as Learning Object Repositories, institutional repositories and latterly Open Educational Resources, have been developed to accommodate collections of learning material that can be used for instructive and research purposes. In order to make this learning material publicly available through internet searching mechanisms, it has to be enriched with metadata. Nowadays, emerging Web 3.0 applications necessitate further enhancing learning resources with semantic knowledge to facilitate conceptually searching. Adding semantic value to educational resources could be achieved by modeling ontologically the metadata profile which describes them. The purpose of this work is to demonstrate the process of enhancing semantically learning resources by developing a new ontology and a prototype system to accommodate them. The designed ontology is originated from a metadata profile which is build according to the guidelines posed by Singapore Framework for Dublin Core Application Profiles. We focus on educational material for teaching courses on Informatics in primary and secondary education in Greece but the methodology and mechanism introduced could be easily adapted in any domain-specific setting.