Dimitrios Tzovaras

RAMCIP: Towards a Robotic Assistant to Support Elderly with Mild Cognitive Impairments at Home

During the last decades the mild cognitive impairments (MCI) as well as the early stage of dementia comprises a societal challenge in the growing elderly population. This fact is highly related to the physical and cognitive decline of aged people, influencing the way they apprehend their environment and, thus, their daily activities. Towards this direction, the “Robotic Assistant for MCI patients at home” (RAMCIP) project, initiated by the European Union, intends to build a service robot that will operate in domestic environments with the aim to proactively and discreetly support older persons and MCI patients. The key component to achieve this goal is the design of a robot endowed with high-level cognitive functions, driven by advanced human and environment perception mechanisms, that will enable the artificial agent to autonomously decide when and how to assist. The paper in hand demonstrates the RAMCIP concept through identified user requirements and provides an overall system description. Additionally, the architecture design of the robotic system is exhibited here, firstly by providing a conceptual analysis and then by further decomposing the identified modules into functional components. The overall architecture envisaged in a user centric manner aiming to convert the real needs of the MCI patients into capabilities of the robotic assistant.

An analysis of personalized web accessibility

This paper presents an experimental study designed to understand the differences of performing personalized Web accessibility evaluations or generic evaluations. We used the WaaT evaluator, which can perform personalized WCAG 2.0 based evaluation. We claim that these tools can be used by an individual to understand if a page is accessible to himself, before wasting time browsing its content. Our study shows that there are considerable differences between the generic and the personalized evaluation results. It also shows that the number of barriers found in accessing pages is in average considerably lower for users with upper limb and cognitive impairments than for those with vision and hearing impairments. Finally, we argue that common accessibility evaluation metrics are inadequate to understand the differences between these user perspectives of the same page. For a particular metric, we show that score differences are diminished and sometimes even mislead the perception of the accessibility quality.

An accessibility assessment framework for improving designers experience in web applications

The current situation for the accessibility assessment of web applications is encouraging, though not sufficient. Many efforts have been made for the development of various tools that carry out the evaluation of web pages. However, their effectiveness is somehow incomplete. In this work we propose a new framework that has adopted the latest web standard and supports a plethora of assessment configurations to meet the user needs.

Using Blockchains to Strengthen the Security of Internet of Things

Blockchain is a distributed ledger technology that became popular as the foundational block of the Bitcoin cryptocurrency. Over the past few years it has seen a rapid growth, both in terms of research and commercial usage. Due to its decentralized nature and its inherent use of cryptography, Blockchain provides an elegant solution to the Byzantine Generals Problem and is thus a good candidate for use in areas that require a decentralized consensus among untrusted peers, eliminating the need for a central authority. Internet of Things is a technology paradigm where a multitude of small devices, including sensors, actuators and RFID tags, are interconnected via a common communications medium to enable a whole new range of tasks and applications. However, existing IoT installations are often vulnerable and prone to security and privacy concerns. This paper studies the use of Blockchain to strengthen the security of IoT networks through a resilient, decentralized mechanism for the connected home that enhances the network self-defense by safeguarding critical security-related data. This mechanism is developed as part of the Safe-Guarding Home IoT Environments with Personalised Real-time Risk Control (GHOST) project.

A Personalized Cloud-Based Platform for AAL Support to Cognitively Impaired Elderly People

Population ageing due to declining fertility rates and/or rising life expectancy is poised to significantly transform our societies in the upcoming years. Although a lot of work has been done in the field of AAL for the creation of ICT solutions that will prolong and support the autonomous living of elderly individuals with cognitive impairments, these solutions are unable to meet all user needs and/or to provide an easy to use mechanism for further extension with new services. The present paper presents a cloud-based solution that provides easy, transparent, personalized and contextualized access to all the supported AAL services to the cognitively impaired elderly end-users and their caregivers by also offering a mechanism for easy registration and integration of new AAL services into the platform.

A Standalone Vision Impairments Simulator for Java Swing Applications

A lot of work has been done lately in an attempt to assess accessibility. For the case of web rich-client applications several tools exist that simulate how a vision impaired or colour-blind person would perceive this content. In this work we propose a simulation tool for non-web JavaTM Swing applications. Developers and designers face a real challenge when creating software that has to cope with a lot of interaction situations, as well as specific directives for ensuring an accessible interaction. The proposed standalone tool will assist them to explore user-centered design and important accessibility issues for their JavaTM Swing implementations.

Parkinson’s Disease Patients Classification Based on a Motion Tracking Methodology

This study demonstrates how a computer based methodology for tracking motor abilities of Parkinson’s disease can be utilized for patient classification and assessment of the Parkinson’s disease severity. The Line Test methodology evaluates the impaired voluntary movement and generates a set of features that describe the motion. A total cohort of 6 control subjects and 37 Parkinson’s disease subjects were recruited and assessed for the test. During the test, a vertical line appears on the screen and the device evaluates patient’s performance by producing features that correlate the motion to the last medication dosage, the line-test position, the line-test reaction time and the line-test total error. A common cohort of 24 Parkinson’s disease subjects (patients that carried out the Line Test more than once) was formed to track the features alterations between repetitions in time. Results evaluation was performed in both cohorts based on information visualization methodology, optimized for the multi-objective dataset. The line-test position and the time from the last medication dosage features were proved to present the major relation to patients’ group formation. Additionally, line-test reaction time and the line-test total error features proved significant between patients’ performance in the common cohort. Study limitations are correlated to the size of the cohort and the time frame of the study. In general, the current practice supports further investigation into using Line Test methodology for addressing Parkinson’s disease severity.

Simulating Heterogeneous Clouds at Scale

In this chapter, a review of existing cloud simulation frameworks is given along with an overview of the recently proposed CloudLightning simulation framework. Moreover, the parallel architecture and parallel implementation details of the CloudLightning simulator are presented along with the characteristics of the supported cloud architectures. These architectures include the traditional centralised approach as well as the Self-Organised and Self-Managed CloudLightning approach. The supported memory, network, and application execution models are reviewed. Furthermore, a recently proposed class of power models for heterogeneous CPU-Accelerator-based hardware is discussed. Finally, large-scale simulations for traditional and Self-Organised and Self-Managed cloud environments are presented and compared.

A framework for inspection of dies attachment on PCB utilizing machine learning techniques

Decision Support Systems are considered as a robust technology able to provide an advantage to several manufacturing companies. As part of the Z-Fact0r EU project Early Stage-Decision Support System, a framework for the inspection of a printed circuit boards (PCB) and the inference of faults, regarding the excess or insufficient glue, is proposed. For the inspection of the PCB, a pixel-based vector of the regions of interest is utilized and several very popular in research community machine learning algorithms are tested on their performance on fault recognition. In order to determine the most efficient and effective classifier, a schema of Monte Carlo simulations for each classification algorithm and set of hyper-parameters was performed. Simulation results show a superiority of the support vector machine (SVM) classifier with polynomial and radial basis function kernels, compared to the rest. The best overall classifier was the SVM polynomial (accuracy: 81.39%, f-measure: 78.72%).

Extraction of Tabular Data from Document Images

In this paper, we propose a heuristics-based method for automatic detection and extraction of tabular data from document images. The proposed approach utilizes page segmentation techniques, along with an OCR engine, in order to acquire the text data and bounding boxes of each word in the document. These elements are then grouped in a bottom-up fashion, based on a series of rules, in order to identify and reconstruct tabular arrangements of data. Based on this methodology, an open source cross-platform tool capable of recognizing the semantic structure of documents containing tabular data has been implemented, thus widening the range of document types than can be successfully converted into alternative accessible formats, suitable for users with visual impairments.