Saeik Firdose employed as an Early Stage Researcher at De Montfort University, School of Computer Science and Informatics within the EU H2020 Marie Sklodowska-Curie ITN ACROSSING project. His research subject includes complex activity modelling and recognition within the smart environment.

Firdose received B.Tech degree in Electronics and Communications from SSITS, JNTU Hyderabad, India(2004-2008), and his M.Tech degree in Information Technology-Networks from Vellore Institute of Technology University, India(2008-2010). From 2011to 2014 he worked in the industry as System Engineer/Application Developer (2011-2014) in telecommunications sector at IBM India Pvt Ltd, Bangalore, India. And also, he worked as a Junior Researcher (2014-2017) with the R&D group SITI, COPELABS, Lisbon, Portugal. COPELABS is an interdisciplinary research unit focused on promoting well-being via pervasive wireless systems. As a Junior researcher, involved in a range of projects that target the users and user-centric networking. The outcomes of some projects are Android applications that provide a way for users to communicate, that allow them to understand their social interactions. His research interests are in the field of Wireless networks, Wireless sensor networks specifically, Social-aware, and user-centric wireless architectures through pervasive and ubiquitous systems including sensors, and mobile devices in the areas such as Context-awareness, proxemics, Health-care, Internet of Things etc.

Sarah Fallmann is employed as an Early Stage Researcher at De Montfort University, School of Computer Science and Informatics within the EU H2020 Marie Sklodowska-Curie ITN ACROSSING project. Her research subject includes behavioural change analysis as well as pattern recognition and decision making, where she uses home automation and body worn sensors.

Sarah has successfully graduated her studies of Mathematics in Science and Technology as well as Biomedical Engineering in 2015 at Vienna University of Technology. Her interest in combining technical knowledge and practical applications led her to take courses in modelling and life sciences. During her studies she took an internship at the University of Leoben, where she increased her abilities by expanding a system developed during the European PinView project for searching in big picture databases. Later on she served internships at the Austrian Institute of Technology. Her research topic was Active and Assisted Living in regards to real world applications. The research was specialized on activity and behaviour recognition using Machine Learning methodologies from unsupervised and supervised nature. Thus, she is aware of chronic diseases, state of the art models in the field of behaviour recognition and Ambient Assisted Living.

Behavioural analysis for personalised self-management of chronic diseases: ESR2 is working on a personalized self-management system for chronic disease patients. This includes the building of a behaviour analysis model detecting daily behaviour with the latest wearable sensors and internet of things including among others motion sensors, door sensors and fitness trackers. The combination of different sensors will make the system adaptable to different users and help to meet the demands individually. The model will include vital signs and human behaviour to detect changes concerning chronic disease problems which will help patients to self-manage their health and lifestyle, and access their data anytime. To achieve this goal a behaviour data model will be constructed including data mining and pattern recognition algorithms. Later on, decision making support mechanisms will be used to transport personalized important information to the users and care givers.

Ismini is an Early Stage Researcher at the EU H2020 Marie Sklodowska-Curie ITN ACROSSING project and a PhD candidate within the School of Computer Science and informatics at De Montfort University. She received her BSc in Informatics and Telecommunications Engineering from the University of Western Macedonia in her home country (Greece) and her MSc in Computer Science from Delft University of Technology. During her studies at TU Delft she focused on Interactive Intelligence and Human-Computer Interaction. Her research topic was the development of an intelligent robotic ePartner to support music therapy and foster social participation in elderly people affected by dementia.

She has also been a Visiting Student Researcher at Massachusetts Institute of Technology (MIT) Media Lab where she worked on the design and evaluation of eHealth and Wellness projects. A few of those include her research on management of chronic diseases. Such as an application for blood pressure management using wearable sensors, were the design won the Health and Wellness Innovation award from the university. As well as a platform for children’s asthma management using persuasive technologies, gamification and prototype sensors for drug intake measurements and adherence. Her research interests include design and personalization of eHealth systems, intelligent agents for management of chronic diseases, privacy and security in assistive technologies, human-computer interaction, ontology-based knowledge engineering, value-sensitive design and situated cognitive engineering.

ESR 3 - Privacy-preserving, Standard-based Wellness and Activity Data Modelling and Management within Smart Homes: The huge amounts of data available through smart home sensors are transforming Ambient Assisted Living. However, in today’s smart home environments there is a lack of mechanisms to integrate the heterogeneity of data and enable inhabitants to perceive and control the data generated by applications, sensors and smart devices in their home. This project will create comprehensive security and privacy models for IoT intensive systems in smart homes. It will investigate how smart home systems can make autonomous decisions about the home, and what role the security and privacy models can play in this context. It will also provide methods for access control and reduced sensitivity in user-generated data. In order to develop these methods it is necessary to also examine the information flow and categories of data, the user interaction within the smart home, and level of autonomy.

Giovanni obtained a MRes. in Advanced Engineering Robotics at Bristol Robotics Laboratory, "University of West England", Bristol (UK). He obtained a MSc. in Biomedical Engineering, with focus on bioinformatics, from "Università Campus Bio-Medico di Roma", Rome (Italy) in 2011. Between the Master by Research and the Master of Science he worked as technical support assistant in an industrial softwares company in Italy, studying and testing industrial automation systems. He obtained a BSc. in Biomedical Engineering from "Università Campus Bio-Medico di Roma", Rome (Italy) in 2009.

Sensors and processing capabilities of modern embedded computers, including mobile, wearable, and implantable devices, can be used to recognise a user's situation and environment continuously, providing context awareness. As sensor modalities and data bandwidth increase, context recognition applications diversify and continuously run statistical pattern recognition algorithms. However, the pattern recognition does not require a constant processing load at all times, i.e., to repeatedly apply sliding windows over equidistant sampled data and sequentially iterate recognition algorithms over the windows. Instead, sampling and processing at relevant moments will be sufficient to obtain acceptable accuracy, where the system could sleep in low power states during the remaining time. To realise a dynamic, on-demand context learning and recognition approach, it is essential to investigate sparse representations and suitable process control strategies that can optimise between accuracy, energy, and latency, while being configurable to support varying services requested by applications.

Affiliation is Universität Passau, ACTLAb Research Group.

Umar Ahmad is an Electrical Engineer from University of Engineering & Technology, Taxila (Pakistan) and has Master of Sceince degrees in Embedded Systems from Technical University of Eindhoven, The Netherlands and in Information & Communication Technology from Technical University of Berlin, Germany.

His Previous research work inculdes designing a prtotype for BLE based indoor localization system. His research interest includes Internet-of-Things and its applications in Health & Wellbeing and Home Automation. In March 2017 he joind University of Passau as Marie-Curie fellow in Computer Science starting his PhD investigation on the use of werable IoT for health and Well-being for the elderly".

His project investigates the potential of novel sensors for healthcare applications. Areas of interest include, but are not limited to: ultra-low power radar, thermal wall/ceiling integrated sensors, body-worn printed sensors and Wearable IoT. Transducer design variants and signal processing methods will be considered to extract information representing health-related behaviour. An incremental sensor design and evaluation methodology is planned, where prototypes are first evaluated with healthy individuals and later on with patients. The activities are embedded in a research team.

José Gabriel is a Telecommunication Engineer from Universidad Católica Andrés Bello (Venezuela), and Master of Science in Electronic System Engineering from Technical University of Madrid. He has experience as Project Manager and Development Director in different Research and Start-ups Companies and part-time professor of Embedded System Design at the Universidad Católica Andrés Bello.

His previous research includes:

-Design and Development of an Indoor location and Guidance System for blind people, using Wireless Sensor Network (WSN). (Prototype implemented in a Metro de Madrid Station).
-Design And Development of a real time inventory tracking and cold chain control in industries using RFID and WSN technologies (Prototype implemented at the national distribution centre of a Major Fast Food Restaurant Company)

He is part the Life Supporting Technologies group from The Technical University of Madrid as a research for the ACROSSING Project.

ACROSSING Project Description

ESR06: Towards personalized in-home self-management and intervention for patients with Chronic Diseases.

EU health care services are striving to change the care paradigm from in-hospital patient care to home centred care.

Personalised self-management care systems have been proven to considerably benefit patients who have experienced different chronic diseases. However, countless patients still do not benefit from proper personalised in-home care, usually because of lack of resources, which leads to the occurrence of frequent relapses and visits to the hospital.

This ESR individual project aims to develop underlying technologies for personalised home-based self-management and intervention for chronic diseases using novel HCI techniques and user-centred design principles mainly for patients with Chronic Diseases.

Nikolaos Liappas is an Early Stage Researcher working for the group of Life Supporting Technologies (LST) within the Universidad Politécnica de Madrid (UPM) under the scientific supervision of Associate Prof. María Fernanda Cabrera-Umpiérrez. His research addresses the topic of virtual reality enabled adaptive interfaces for cognitive rehabilitation and training, targeted for People with Mild Cognitive Impairment. As a graduate from National and Kapodistrian University of Athens, with a Ptychion degree in Informatics and Telecommunications, he has undertaken several placements within leading organisations such as Vodafone Greece and the European Commission. He has over of two years working experience in the field of computer science, in both IT companies and research laboratories, within different international environments including six countries. His final thesis will lead him towards a PhD degree in Biomedical Engineering.

Non-pharmacological interventions have proved to be more effective than pharmacological drug alternatives mediations regarding Dementia and AD´s disease. Managing the quality of life of individuals with MCI is challenging and risky, especially for the family related caregivers. The most common issue of people with MCI is that they are unable to perform instrumental activities of daily living (IADL), including simple things such as cooking meals, bathing, etc., and more memory dependent functions like managing their finances. In addition, caregivers sometimes must cope with even more stressful situations when the individuals suffer from anxiety or apathy. At this point, emerging technologies in combination with assisted living can be the key element to improve the quality of life and extend/prevent this chronic disease cycle. Assisted living is in favor of providing great support to MCI people due to the fact that it can support the needs of those people.

Nikolaos Liappas is doing research on identifying the needs of MCI people and proposing interventions based on emerging technologies.

Current proposed Interventions include: Virtual Assistant Interface to support the following needs: social company, independence from caregivers and clinics (since most patients prefer to stay at home), and most importantly to instruct them in daily activities.

Virtual Reality for Cognitive Training: Improve cognitive performance and also simulate virtually daily activities such as cooking, bathing, etc.

The expected results include: MCI / AD patient and new interface/interaction technology analysis. Cognitive models and VR based monitoring and interaction mechanisms. VR interaction and cognitive performance analysis algorithms and adaptation methods. System usability and performance validation.

I am working on deep reinforcement learning for dialogue systems. Before joining Acrossing project I worked at IBM Watson as a core software developer and Google as Technical Analyst. I graduated from Sabanci University with Bsc. Electronics and Msc. Computer Science.

Project will mainly focus on dynamic dialog systems, capable of practically be deployed on real world applications, through avatars to increase the quality of smart home avatars designed for older persons with cognitive impairments.In our proposal we will be addressing the issue of dialogue creation tool that is needed to create samples to train reinforcement learning and also focus on more sample-efficient and adaptive reinforcement learning methods. Since people with cognitive impairments tend to change their behavior in course of time and their preferences, it is also crucial to develop a reinforcement learning algorithm which will take into account temporal changes on problem space for same user. (Action, state, reward) Automated creation of dialogues and more efficient and intelligent systems would make it possible to use dynamic dialogue systems to be used in smart home applications for people with dementia.

Affiliation: Austrian Institute of Technology Vienna

Deepika completed her Bachelor’s and Master’s studies in Computer Science Engineering in India. The use of information technology in health sector applications and towards enhancement of people’s lives has been her major field of interest. Previously, during her master’s, she worked on project entitled “Early detection of the Malignant Melanoma” at Malaviya National Institute of Technology Jaipur, India and later on joined the same organization as a lecturer cum research fellow.

Currently, She is an early stage researcher within the ACROSSING and employed as a Junior Scientist at Austrian Institute of Technology (Vienna, Austria). Her research focuses on an open Smart Home (SH) platform which should be extensible, adaptable and configurable to different application scenarios and it must be capable of intelligently integrating data from users, different sources, software components and objects. She is focusing on ambient intelligence and context awareness techniques to develop modules which should be adaptable to user needs and requirements.

Georgios is ESR10 in the ACROSSING project and his part evolves around the use of first person view cameras for activity monitoring. In the past, he has worked on personal user profile creation through a smartphone application based on Android for his undergraduate Diploma Thesis. Following, his work shifted on the computer vision domain where he developed a system for text detection in the wild. This work concluded a MSc Thesis on the Department of Electrical and Computer Engineering in Democritus University of Thrace, Greece. Georgios' current project combines computer vision and behavioral understanding and as such, it is a perfect mixture of his background. What motivates Georgios is the development of technology in a manner that is human-centered and the ACROSSING project offers the possibility to shape novel technology towards this goal.

The project of Georgios comprises the embedding of a first-person wearable camera in a Smart-Home environment. The end goal is to achieve real-time understanding of the activities performed by the camera wearer, from their own perspective. Ultimately, this information will lead to the identification of behavioral patterns connected to the diseases they suffer from, in order to detect signs of progression or improvement. The objectives of ESR 10 in the ACROSSING project consist of the development of an object detection system that operates in controlled environments through a egocentric wearable camera. Analysis of the correlations between the objects will lead to activity detection. Furthermore, the integration of different sensors to enhance the comprehending capabilities of the system will be investigated. Context recognition utilizing video information is not fully developed against complex scenarios and external sensors will provide more information towards such distinctions.

Georgios is hosted at Noldus Information Technologies in Wageningen the Netherlands and pursues his PhD degree at Utrecht University.

I am (Rick van Veen) an Early Stage Researcher (ESR) within the EU H2020 Marie Sklodowska-Curie International Training Network (ITN) ACROSSING project and is stationed at the Philips Research site in Cambridge in the UK. I received both my BSc and MSc in Computing Science from the University of Groningen, where I specialized in Intelligent systems. Using this gained knowledge, I will now tackle the following problem within the ACROSSING project.

With the aging human population, mental illness is becoming an increasingly bigger problem. Early detection of, e.g., mild cognitive impairment (MCI) and dementia, is of utmost importance to slow down the progression of these conditions.

The project includes the development of a novel method to use smart home (SH) and body worn sensors to detect the activity profiles of the older population. A data mining algorithm is developed to detect changes in the behaviour and the way daily activities are performed. These daily activity data are used in combination with results from validated cognitive test to build a hybrid CI risk assessment method.

Ultimately, all these aspects are combined in a system that can assist healthcare specialists to make better risk assessments, as early as possible.

Mohammad Reza is Early Stage Researcher based at Vienna University of Technology, Austria. He is working in the Vision for Robotics (V4R) lab developing advanced algorithms to enhance robots' visual perceptual system.

Mohammad Reza is a young and curious engineer that completed his Bachelor's studies in Electronic and Telecommunication Engineering in Italy. Recently he obtained a double Master degree in Advanced Robotics, gaining a 360 view of the different fields that characterizes robotics' interdisciplinarity. His professional interest is focused on Artificial Intelligence algorithms and their practical applications in robotics to enhance humans well-being.

ESR12 - Robot Gaming Strategies to Enhance Older Adult's Wellbeing and Quality of Life: The use of an active agent, such as a Service Robot (SR), in smart-home based assisted living applications opens to a whole new world of possibilities. The primary function a SR should be able to accomplish in order to provide assistance to people is to sense and understand the surrounding environment. The task of recognizing object in real-time in the domestic setting, with its plethora of categories and their inter-class variance, it is highly challenging and still remains an unsolved problem. The goal of this project is to design and implement an effective algorithm providing abstract knowledge about the objects populating the home environment and to use this knowledge to address user's needs. Finally, gaming strategies will be used to include the assisted person in the algorithm's workflow, not only as end-user, but also also a source of knowledge.

Anastasios holds a BSc in Electrical & Computer Engineering from Worcester Polytechnic Institute (WPI) and a MSc in Electrical & Computer Engineering from Northeastern University.

He has worked as a Research Assistant at Toyota InfoTechnology Center, USA; conducting research regarding vehicular communications and Dynamic Spectrum Access (DSA). He has also worked as an Advance Development Engineer at Bose Corporation, developing algorithms for active noise cancellation and engine harmonics enhancement in vehicles with Continuous Variable Transmission (CVT) system.

In this project, the use of wireless acoustic and motion sensor network for daily activity monitoring and recognition will be investigated, developing; a) a smart audio-based sensing network, b) acoustic signal processing methods and activity recognition algorithms, c) a novel automatic trajectory-based activity detection framework to recognize activity patterns and user profiles, d) associated technologies and a prototype for evaluation by real users.

The research interests include digital signal processing, machine learning and acoustics.

Affiliation: Center for Research and Technology Hellas (CERTH) - Information Technologies Institute (ITI)

Eduardo is Early Stage Researcher hosted by ISOIN, Spain. He enjoys combining his research in adaptive user interfaces with his sportive life. During his career in Telecommunication engineering and computer science, Eduardo was several times national university champion in distinct categories such as kickboxing, swimming and water polo. The close collaboration with a care centre after his graduation sparkled his interest for ACROSSING project. He seeks to use his experience in ICT to overcome the accessibility gaps of digital technology for elderly people.

Due to the aging process elderly people start facing common types of cognitive and physical impairments (visual, hearing and cognitive workload) that increase the difficulty of interaction with new technologies. In this respect, ESR14 will reduce these barriers developing a platform over an infrastructure that accordingly to the user profile, the usage and the input of various sensors, the system will automatically adapt the interface and the content.

Moreover, this infrastructure will be designed to provide interoperability of services, this means that new services can be added to the platform and the interface will still be personalized, reducing the adaptation period of older people and new services. With this user centre design, elderly people will be more confident to use and adopt new technology as it is prepared for them, increasing the levels of usability, value and engagement of older people with new technologies.

Federico Cruciani received his BScEng and MscEng in Computer Science at University of Florence (Italy). From 2006 to 2016 he worked as software project manager at I+ and Orthokey Italia with main focus on Computer Assisted Surgery and Smart Environments for healthcare. He gained experience on IEC-62304 standard for medical device software working on optical tracking, computer vision and inertial sensors for orthopaedic surgery, kinematic analysis, and rehabilitation. In September 2016, he joined Ulster University as Marie-Curie fellow in Computer Science starting his PhD investigation on the use of smart technologies to promote physical activity among older adults.

Project title: The use of connected-health to promote physical activity

Lack or inadequate level of physical activity is a major risk factor in health. Still, however, a significant percentage of the population, particularly among older adults, live a sedentary lifestyle. This project aims to explore the use of connected health technologies in order to promote the conduction of an active and healthy lifestyle. The project’s research objectives include the use of sensors to assess the level of physical activity, the definition of dynamic models for behaviour change and the development of a reasoner able to identify the most appropriate behaviour change strategy to promote physical activity for the specific individual.