Tuesday, 11 Aprilie 2023
Session 1: Sustainable Computing, Chair: Marin Litoiu
Alexandru Iosup, VU University Amsterdam
Massivizing Computer Systems: the Science, Design, and Engineering of Distributed Ecosystems
Abstract:
Our society is turning digital. Science and engineering, business-critical and economic operations, and online education and gaming rely increasingly on the effective digitalization of their processes. For digitalization to succeed, societal processes must leverage efficient computer systems, effectively and efficiently integrated into larger _ecosystems_, managed primarily without application developer and even client input. However successful until now, we cannot take these ecosystems for granted: the core does not yet rely on sound principles of science and design, and there are warning signs about the scalability, dependability, and sustainability of engineered operations. This is the challenge of massivizing computer systems.
We can address this grand, fundamental challenge by focusing on computer ecosystems rather than merely on (individual, small-scale) computer systems. In this talk, we focus on understanding, deploying, scaling, and evolving computer ecosystems successfully. To this end, we define (distributed) computer ecosystems and differentiate them from (distributed) computer systems. We formulate principles and introduce a reference architecture for computer ecosystems supporting diverse workloads – AI/ML, big data and graph processing, online gaming and metaverse, and business-critical and serverless – and diverse resources and backend services across the computing continuum. We synthesize a framework of resource management and scheduling (RM&S) techniques, which we argue should be explored systematically in the next decade. We show early results obtained experimentally, through controlled real-world experiments, long-term observation, and what-if analysis of short- and long-term scenarios using the OpenDC digital twin for datacenters.
This talk is a call to the entire community [1]: there is much to discover and achieve, and to get meaningful, long-lasting results we need to form a community toward holistic improvements of applications, services, and processes, together with the computing infrastructure supporting them.
—
[1] Future Computer Systems and Networking Research in the Netherlands: A Manifesto, 2022. [Online] https://arxiv.org/pdf/2206.03259
Brief Bio:
Dr.ir. Alexandru Iosup is a full professor at Vrije Universiteit Amsterdam (VU), a high-quality research university in the Netherlands. He is the tenured chair of the Massivizing Computer Systems research group at the VU and visiting researcher at TU Delft. He is also elected chair of the SPEC-RG Cloud Group. His work in distributed systems and ecosystems includes over 150 peer-reviewed articles with high scientific impact, and has applications in cloud computing, big data, scientific and business-critical computing, and online gaming and the metaverse. His research has received prestigious recognition, including membership in the (Young) Royal Academy of Arts and Sciences of the Netherlands, the Netherlands ICT Researcher of the Year award, and a PhD from TU Delft. His leadership and innovation in education led to various awards, including the prestigious Netherlands Higher-Education Teacher of the Year. He has received a knighthood for cultural and scientific merits.
Contact Alexandru at A.Iosup@vu.nl or visit http://atlarge.science/aiosup
Ana-Lucia Vȃrbănescu, University of Twente
Towards Zero-Waste Computing.
Abstract:
“Computation” has become a massive part of our daily lives; even more so, in science, a lot of experiments and analysis rely on massive computation. Under the assumption that computation is cheap, and time-to-result is the only relevant metric for all of us, we currently use computational resources at record-low efficiency.
In this talk, I argue this approach is an unacceptable waste of computing resources. I further introduce _performance engineering_ methods and techniques to facilitate zero-waste computing. By means of a couple of case-studies, I will also demonstrate performance engineering at work, proving how efficiency and time-to-result can be happily married.
Finally, I will provide insights and ideas for quantifying computing waste in terms of lack of efficiency, and propose a strategy for system co-design to demonstrate how zero-waste can be achieved.
Brief Bio:
Ana Lucia Varbanescu holds a BSc and MSc degree from POLITEHNICA University in Bucharest, Romania. She obtained her PhD from TUDelft, The Netherlands, and continued to work as a PostDoc researcher in The Netherlands, at TUDelft and VU University in Amsterdam. She is a MacGillavry fellow at University of Amsterdam, where she was tenured in 2018 as Associate Professor. Since 2022, she is also Professor at University of Twente. She has been a visiting researcher at IBM TJ Watson (2006, 2007), Barcelona Supercomputing Center (2007), NVIDIA (2009), and Imperial College of London (2013). She has received several NWO grants (including a Veni grant) and she is co-PI for the GraphMassivizer EU project.
Ana’s research stems from HPC, and investigates the use of heterogeneous architectures for high-performance computing, with a special focus on performance and energy efficiency modeling for both scientific and irregular, data-intensive applications. Her latest research focuses on zero-waste computing and systems co-design.
Session 2: Smart Systems and their Engineering, Chair: Anca Daniela Ionita
Marin Liţoiu, York University
AI and System Self-adaptation: Results, Challenges and Risks
Abstract:
Self-adaptation in computing refers to the property of the software systems to monitor and analyze themselves and their environment, to detect or predict deviations from their specified goals, plan and execute actions that correct those deviations. Prediction and planning in self-adaptation are based on models that are identified at run-time. The models are built by measuring the control inputs, disturbances, and outputs of the adapted system and fitting the data into a function. In this presentation, we will discuss the role of Artificial Intelligence models in achieving self-adaptation, their promises, challenges, and side-effects. We will present our experience in building Artificial Intelligence models that predict resource usage and the impact of changes. These models enable look-ahead optimization as a means to achieve performance with fewer resources and less energy, thereby enabling sustainable and responsible computing. We will also discuss the risks of self-adaptation, the possible side effects, and ways to mitigate them.
Brief Bio:
Marin Litoiu is a Professor of Software Engineering in the Department of Electrical Engineering and Computer Science and in the School of Information Technology, York University. He is also a Fellow of the Canadian Academy of Engineering. Dr. Litoiu leads the Adaptive Software Research Lab and focuses on making large software systems more versatile, resilient, energy-efficient, self-healing and self-optimizing. His research won many awards including the IBM Canada CAS Research Project of the Year Award, the IBM CAS Faculty Fellow of the Year Award for his “impact on IBM people, processes and technology,” three Best Paper Awards and two Most Influential Paper Awards. Prior to joining York University, Dr. Litoiu was a Research Staff member with the Centre for Advanced Studies in the IBM Toronto Lab where he led the research programs in software engineering and autonomic computing. He received the Canada NSERC Synergy Award for Innovation in recognition for these collaborative university/industry activities. He was also recipient of the IBM Outstanding Technical Contribution Award for his research vision on Cloud Computing. Dr. Litoiu is one of the founders of the SEAMS Symposium series—ACM/IEEE Software Engineering for Adaptive and Self-Managing Systems. He has been the General Chair of SEAMS in 2013 and 2019, ACSOS 2023, ICPE 2025, Program Chair of SEAMS, ICPE and CASCON and serves on the steering committees of SEAMS and CASCON. Dr. Litoiu is also the Scientific Director of “Dependable Internet of Things Applications (DITA),” an NSERC CREATE program.
Ada Diaconescu, Telecom Paris
Multi-scale Architectures for Sustainable and Scalable Socio-Technical Systems
Abstract:
Smart technology is progressively embedded into our living environments and inter-meshed with essential infrastructures and human organisations. The sustainability and scalability of the emerging socio-technical systems(-of-systems) becomes a major and rather urgent question.
Most large-scale systems – from organisms and human organisations to distributed algorithms and control systems – feature a ‘hierarchical’ structure. Here, higher-levels observe and control lower-levels, in a recursive divide-and-conquer approach. However, the default concept of ‘hierarchy’ often implies top-down authority, reductionism and tree-like topologies (centralised at the top). We argue that such features are neither necessary nor viable for all systems; and investigate a wider paradigm of “multi-scale” architectures, to propose alternative designs.
Our research aims to identify the key aspects of multi-scale architectures that ensure their sustainability and scalability. Initial findings include inter-scale information abstraction, feedback, timing and flow distribution across system resources. We investigate the main alternatives for implementing these generic aspects, depending on the requirements and constraints of each system (e.g. need for reactivity, accuracy, stability; resource availability and distribution). System engineers can rely on such reusable findings to select and customise system architectures that are best suited to their particular contexts. Applications concern all socio-technical systems(-of-systems) that need to coordinate a large number of internal processes – e.g. from smart cities, traffic and power grids, all the way to bottom-up organisations and social networks.
Brief Bio:
Ada Diaconescu fulfills the role of Assistant Professor (tenured) at Telecom Paris, Institut Polytechnique de Paris, since 2009. She is the head of the Autonomous and Critical Embedded Systems (ACES) team in the Computing and Networks (INFRES) department. Her research interests include self-organising complex adaptive systems, multi-scale feedback architectures, and the impact of technology in socio-technical systems.
Having received a Ph.D. degree from Dublin City University in 2006, she pursued several Post-Doctoral Research positions at the University of Grenoble, Orange Labs, and INRIA Rhone Alpes, from 2005 to 2009. She spent a sabbatical at Leibniz University, Hanover, 2016-2017.
She has co-authored a Springer book on Autonomic Computing in 2013. She was a PC Co-Chair of IEEE SASO in 2014, a General Co-Chair of IEEE ICAC in 2015, a General Chair of IEEE SASO in 2017, and a Co-Organizer of two Dagstuhl seminars in 2015 and 2018. She is Steering Committee Co-chair of the IEEE Conference on Autonomic Computing and Self-Organising Systems (ACSOS), since 2019. She is co-organising a workshop on Self-Improving System Integration (SISSY), since 2019; and a workshop on Sustainable and Scaleable Self-organisation (SaSSO), in 2023.
Miruna Elena Iliuță, Mihnea Alexandru Moisescu, University Politehnica of Bucharest
Risk Assessment for Digital Twins in Systems Medicine
Abstract:
Digital twin technology has emerged as a promising tool in Systems Medicine for improving diagnosis, treatment, and disease prevention. However, the adoption of Digital Twin technology in healthcare also poses risks that need to be assessed and managed. This paper presents an abstract for a risk assessment study of digital twin technology in systems medicine. The study aims to identify potential risks associated with the use of digital twins in healthcare and develop a framework for risk assessment and management. The proposed framework will consider various aspects of digital twin technology, including data privacy and security, system interoperability, model accuracy, and ethical considerations. The study will also propose strategies for mitigating the identified risks and ensuring the safe and effective use of digital twin technology in systems medicine. The findings of this study will be useful for healthcare organizations and other stakeholders in developing risk management strategies and guidelines for the use of digital twins in healthcare.
Brief Bio:
Miruna – Elena ILIUȚĂ is a teaching assistant at University Politehnica of Bucharest, Faculty of Automation and Computers Science, Department of Automation and Industrial Informatics. She followed her bachelor’s and master’s studies at the Faculty of Engineering in Foreign Languages, and is currently completing her doctoral studies, wanting to pursue a career in the university field. She carries out her teaching activities in Project Management and Software Engineering, also being involved in the coordination of diploma projects.
Session 3: Software Engineering for Smart Systems, Chair: Dana Petcu
Dan Ionescu, University of Ottawa, Canada
Blending Algorithmic and Artificial Intelligence Platforms for Higher Precision and Shorter Data Sets
Abstract:
Using algorithmic approaches to find answers to questions in regards to a specific domain such as image processing can reduce the search space and thus the computational burden, however the precision of the solution search depends very much on the existing models and data sets. Recent years have opened, due to the new computer technology advances, new perspectives in using Artificial Intelligence empirical methodology due to the revival of the Frank Rosenblatt’s Perceptron, called nowadays Deep Neural Network (DNN). This approach proposes to learn the solution to the task from data using Machine Learning methods such as recurrent neural networks (RNNs) i.e., long short-term memory (LSTM), gated recurrent units (GRUs) or others.
The application of DNN approaches to finding the solution in the domain of expertise, proved being better in dealing with models’ subtleties.
The challenge though is in the amount of data needed to train the model and also the computational time needed to solve completely the question (s).
However, there are domains in which developed algorithms succeeded to implement highly efficient methods capable of determining with a high precision the process model.
Therefore, combining, or blending specific algorithms and artificial intelligence platforms has the potential to lead to higher precision and shorter data sets. Algorithmic platforms are usually designed to solve specific problems using a set of rules, while artificial intelligence platforms use machine learning algorithms to learn from data and improve their performance over time.
By blending these two approaches, it is possible to leverage the strengths of each to create more accurate and efficient solutions.
At NCCT the research is directed on using Finite State Machines (FSM) to automate the blending of the time-series algorithms with GRU, Reinforcement Learning directed via Transformers.
The research in this field is encompassing the following directions;
1) Hybrid models: One approach is to create hybrid models that combine the rule-based (Expert System) approach of algorithmic platforms with the data-driven approach of AI platforms. As in the era of Expert Systems we will create a hybrid model based on a set of rules for pre-processing data set that is process by a DNN.
2) Feature engineering: In this approach the algorithmic platforms such as for example time-series are used for performing filtering. This can help improve the performance of AI models by reducing the amount of data to be used for training
3) Multimodal Learning: Multimodal learning is a technique that combines multiple models to achieve higher accuracy than any individual model. By blending algorithmic and AI models, one can create more diverse ensembles that are more robust to different types of data and noise.
As well as in the Expert System approach the final DNN will be capable to select the Neural Network Architecture which is the best obtained by the Reinforcement Learning Platform.
Brief Bio:
Dr. Dan Ionescu is a Professor in the School of Information Technology and Engineering (SITE) more recently known as the School of Electrical Engineering and Computer Science (EECS), University of Ottawa, Ottawa, Ontario, Canada.-known also the Capital City University, one of the top university in Canada having more than 50,000 students. Some worldwide accepted classifications show the University of Ottawa among the first 111 universities in the world.
Dr. Dan Ionescu, received his Dipl. Eng. and Dr. Eng. degrees in Control and Computing from the Polytechnic Institute of Bucharest, Romania, and Dipl. in Mathematics from the University of Timisoara. He is a Registered Professional Engineer in the Province of Ontario, Canada, and a Senior Member of various scientific organizations such as IEEE and ACM.
Dr. Ionescu career started as a research engineer in the Research Institute for Hydro-Equipment of Timisoara, Romania, then as Assistant Professor in the Faculty of Electrical Engineering at the Polytechnic Institute of Timisoara, Romania, a professor with the University of Dusseldorf, Germany, and a Professor with th eElectrical and Computer Engineering Department of the University of Ottawa, Canada. At the university of Ottawa Dr. Ionescu has dedicated his activity to teaching and research.
He is the Director of the Network Computing and Control Technologies (NCCT, www.ncct.uottawa.ca) research laboratory since 1999, and previously the Director of Machine Intelligence research laboratory. At the University of Ottawa he has been the Director of Ottawa Carleton Institute of Electrical and Computer Engineering (OCIECE, www.ociece.ca) and served also as Director of Computer Engineering from 1996 to 2000. Dr. Dan Ionescu is a senior member of various IEEE, IFIP, SIAM, and IFAC groups. Dr. Ionescu served in various IEEE and ACM Conferences on different functions from General Co-Chair to Scientific Advisor and to Technical Committees. He was invited as a distinguished Keynote Speaker in many IEEE and ACM conferences.
His research at the University of Ottawa spanned over a few domains such as Artificial Intelligence, Machine Vision, Distributed Computing and Network Control. His contributions to Expert Systems, Image Processing, Temporal Logic, Discrete Event and Real-Time Systems materialized in a series of papers and edited books.
Sergiu Dascălu, Pengbo Chu, Levi Scully, Charlotte Moreland, Nathan Bertram, Dustin Hurtz, University of Nevada,
A VR Simulation for Mining Engineering Education
Abstract:
Brief Bio:
Sergiu Dascalu is a Professor in the Department of Computer Science and Engineering at the University of Nevada, Reno (UNR), which he joined in July 2002. He received a PhD degree in Computer Science from Dalhousie University (2001), Canada, and a Master’s degree in Automatic Control and Computers from the Polytechnic of Bucharest, Romania (1982). At UNR he is also the Director of the Software Engineering Laboratory (SOELA) and the Co-Director of the Cyberinfrastructure Lab (CIL). He has worked on multiple research projects funded by federal agencies such as NSF, NASA, and DoD-ONR. He has advised 15 PhD and over 60 Master students. He received several awards, including the 2011 UNR Outstanding Undergraduate Research Faculty Mentor Award, the 2011 UNR Donald Tibbitts Distinguished Teacher of the Year Award, the 2014 College of Engineering Faculty Excellence Award, and the 2019 UNR Vada Trimble Outstanding Graduate Mentor Award. He has published over 50 journal articles and over 200 conference papers. He is a Senior Member of the ACM.
Irina Nedelcu, Amazon EU Sarl, Anca Daniela Ioniţă, University Politehnica of Bucharest
Artificial Intelligence Meets Modeling
Abstract:
One of the current challenges in Model Driven Engineering is to get advantage of artificial intelligence techniques for improving the modeling tools and reaching a broader adoption in practice. This may be obtained with modeling bots, model recommenders, model reviewers, semantic analysis, which may be integrated into software environments to make developers’ tasks easier. This presentation also provides an analysis of examples where artificial intelligence is applied in conjunction with the standard Unified Modelling Language (UML), for automatic classification of diagrams, recognition of semantic elements, or similarity assessment. We present several experiments of UML diagrams’ classification based on machine learning and discuss how the outputs may lead to wrong conclusions.
Brief Bio:
Irina Nedelcu is a Software Development Engineer at Amazon Luxembourg, supporting Last Mile products. She joined Amazon as a Software Development Intern and previously started her career with Deloitte Digital Romania as a Java Developer. Irina Nedelcu received her B.S. (2019) in systems engineering and the M.S. (2021) in computer science from University POLITEHNICA of Bucharest. She is actively engaged into projects falling into applied science that combine software development and artificial intelligence.
Liliana, Dobrică, Mălina Stanciu, University Politehnica of Bucharest
Ontology-based Analysis Tools at Early-stage Development of Cyber-physical Systems
Abstract:
Today’s cyber-physical systems (CPS) are becoming more complex and must follow recommendations of certification standards for safety-critical domain to support guaranteed performance. CPS development include quality analysis activities at early stages to discover and identify possible problems including lack of predictability or unsafe interactions among various system components of diverse types that can lead to hazards and cause accidents. Many analysis methods tend to be more ample and challenging due to the system structure description that is modeled inadequately. Many times, the analysts waste time and make mistakes due to lack of comprehensive, accurate and relevant information. Tools assistance and techniques are necessary to improve and ease the analysis. Such tools may provide guidance to the analysts through the analysis process. This guidance knowledge aims at avoiding doubts and helping the analyst to perform a complete and time-efficient analysis. There is a need for approaches that achieve better results with techniques that make the systems smarter or are able to automate the analysis, such as ontologies. Ontologies formalize domain knowledge to make it explicit, asserted and inferred, and accessible in a representation that can be used by computing machines and humans. An important feature of an ontology is that it is verifiable for inconsistencies and gives confidence in using the tool. A smart tool that integrates an ontology makes the analysis more systematic, automatic and guided. The use of ontology results in a more trustable tool.
Brief Bio:
Liliana Dobrica is professor at University Politehnica of Bucharest, Faculty of Automation and Computers Science, Department of Automation and Industrial Informatics.
Session 4: Smart Applications, Chair: Mihnea Moisescu
Cristina Luca, Amirali Shirazibeheshti, George Wilson, Anglia Ruskin University
Empowering Healthcare Business Management Systems by AI to facilitate decision making
Abstract:
Ensuring that medicines are prescribed safely is fundamental to the role of healthcare professionals who need to be vigilant about the risks associated with drugs and their interactions with other medicines (polypharmacy). One aspect of preventative healthcare is to use Artificial Intelligence to identify patients at risk using big data analytics. The aim of this research was to identify groups of patients at high risk of polypharmacy using AI techniques. A database of 300,000 patients from a UK based healthcare provider was analysed and processed. The technique was implemented into a healthcare management system that easily and automatically identifies groups at risk in a few seconds or minutes as opposed to hours or days that the manually intensive process normally takes. This enables timely clinical intervention by healthcare professionals and will have substantial cost-saving benefits for medical businesses.
Brief Bio:
Dr. Cristina Luca (BSc, MSc, PhD, FHEA, MBCS) is an Associate Professor at Anglia Ruskin University where her teaching includes programming, software engineering and web application development. Prior to her academic career she worked for several years in industry as a senior software engineer. Her research interests are in the semantic web and the integration of ontologies in software applications utilising machine learning. She has co-authored several textbooks and has peer reviewed papers on semantic query languages, linked data, artificial intelligence and formal languages. Cristina has also been involved in a number of income generating projects through knowledge transfer schemes and other sources.
Lăcrămioara Stoicu-Tivadar, University Politehnica Timișoara
HosmartAI – Frame for a Common Open Integration Platform Supporting the Benefits of Incorporating Digital Technologies in the Healthcare System
Abstract:
The presentation is a summary of the EU Horizon financed HosmartAI project and points out tasks related to the FAIR data concept implementation and the European Federation for Medical Informatics team’s work during the project.
The project involves 24 partners from 12 countries, is financed with 10 mil Euros and its duration is between 2021-2024. HosmartAI will create a common open Integration Platform with the necessary tools to facilitate and measure the benefits of integrating digital technologies (robotics and AI) in the healthcare system. A central hub will offer multi-faceted lasting functionalities (Marketplace, Co-creation space, Benchmarking) to healthcare stakeholders, combined with a collection of methods, tools and solutions to integrate and deploy AI-enabled solutions. The Benchmarking tool will promote the adoption in new settings, while enabling a meeting place for technology providers and end-users. The approach will guarantee the integration of Digital and Robot technologies in new Healthcare environments and the possibility to analyze their benefits. Eight Large-Scale Pilots will implement and evaluate improvements in medical diagnosis, surgical interventions, prevention and treatment of diseases, and support for rehabilitation and long-term care in several Hospital and care settings.
The EFMI team will develop an implementation guide of the FAIR data maturity model oriented to facilitate adoption and to infer the implementation profile, promoting the development of an implementation profile of FAIR principles in HosmartAI to drive the application of AI.
Brief Bio:
Lăcrămioara Stoicu-Tivadar – Professor University Politehnica Timisoara – Faculty of Automation and Computers, member of the International Academy of Health Sciences Informatics (IAHSI), former president of the European Federation for Medical Informatics (2018-2020), former vice-president of the International Medical Informatics Association representing Europe (2020-2022), IEEE Senior member, coordinator of the UPT Master in Healthcare Information Systems since 2009, chair of the Education Committee of the UPT Senate. She acts as EFMI coordinator for the EU Horizon 2020 project HosmartAI – “Hospital Smart development based on AI” (2021-2024), is MC for the COST project Net4Age friendly, co-leader of WG3 – Digital solutions and large-scale sustainable implementation and in the COST project Good Brother is actively participating in WG3 – Audio- and video-based AAL applications. (2021-2024). She is PhD Coordinator for thesis that contribute with research results in the IT and medicine areas.
Marc Frîncu, Nottingham Trent University
The Rise of Responsible AI to Balance the Potential Bias in Truly Smart Grids
Abstract:
The recent advances in AI promise to create the first truly smart grid systems. However comes at a price, as biased can be unvoluntarily introduced creating a series of problems around what we know as responsible AI. Already governments and entities around the world are considering regulations to protect against this new threat. In this context the industry sector and academia focus on explainable AI as a first step in ensuring unbiased results for their use cases. This talk introduces these concepts and presents initiatives and examples around smart grids.
Brief Bio:
Marc Frîncu – Senior Lecturer at Nottingham Trent University, UK. I previously worked in a large US DoE Smart Grid project in the area of Los Angeles focusing around automation and machine learning. I also managed as PI several R&D projects revolving around smart grids with a focus on tech transfer and renewables.
Wednesday, 12 Aprilie 2023
Session 5: Responsible AI, Chair: Marin Litoiu
Vio Onut, IBM Canada
Deepfake – Security and Societal Challenges
Abstract:
With the exponential advances in AI, the potential reuse of these technologies to enhance cyber attacks is a reality that security professionals face. Deepfake is a technology that allows altering someone’s appearance in a video. It is easy to see how this irresponsible use can augment a social engineering attack making it more credible. This talk is meant to highlight some of the challenges and security implications for businesses and people facing the malicious use of this technology.
Brief Bio:
Iosif-Viorel (Vio) Onut is passionate about accelerating curriculum and product innovation through R&D. In the past decade, he has managed more than 150 research projects involving 35 universities, led by over 90 professors for over 360 students and over 330 IBM staff. He specializes in cybersecurity and cybercrime. He finished his Ph.D. in 2008 at the University of New Brunswick, specializing in network security, and has worked in security for the past 20 years. Vio holds multiple positions, he currently is Co-Director at the uOttawa-IBM Cyber Range; Adjunct Professors at the University of Ottawa; and Senior Manager, R&D Strategy at IBM Centre for Advanced Studies Canada.
Gheorghe Căpățână, Moldova State University
Intelligent Software for the Research Domain: Mental and Behavioural Disorders in Epilepsy
Abstract:
The diagnosis of mental and behavioural disorders in epilepsy (MBDE) represents important issues in the treatment and recovery of patients with MBDE.
MBDE research has been carried out for 20 years, is assisted by the artificial intelligence and intended for a better understanding of the triggering mechanisms of the MBDE and for to avoid possible misunderstandings in the triggering of these mechanisms.
Our team of researchers, assisted by artificial intelligence means, studies MBDE remissions to understand their working mechanisms and to propose innovative methods to combat MBDE.
We note that currently the epileptologist psychiatrist of our team of researchers, Alexandru Popov, has obtained over 175 remissions in patients with MBDE. The team’s research favors more and more frequent new remissions.
In the presentation will be exhibited:
- Description of the research domain MBDE
- Structure of the knowledge base of the research domain MBDE
- The metric spaces for the research domain MBDE
- The distance tables between MBDE diagnoses
- The similarity tables of MBDE diagnoses.
Some results are for the first time, they were highly appreciated at some International Innovation Fairs (including in Timișoara) and at some Book Exhibitions.
The intelligent software products were developed by applying the Family Oriented Programming of Aplications (FOPA) methodology.
The methodology started in 1972 and was applied in some economic, management and scientific units in the Republic of Moldova and abroad.
Were developed 5 doctoral theses applying the FOPA methodology.
Brief Bio:
Gheorghe CĂPĂȚÂNĂ, University Professor, Doctor of Engineering, graduate of the Faculty of Mathematics and Cybernetics (State University of Chisinau, 1970), Doctor of Engineering (Polytechnic University, Bucharest, 1995).
Holder of positions: computer scientist (Institute of Mathematics and Computer Science, Academy of Sciences of Moldova); Head of laboratory, Head of department (Technical-Scientific Association ”Information Technologies and Systems”, the Agro-Industrial Committee of the USSR) at the same time, member of the Section “Problem Oriented Complexes” (Council of Main Builders, Intergovernmental Commission for the Computing Technique of Socialist States); main specialist (Higher Attestation Commission of the Republic of Moldova); Dean (Cooperative-Commercial University of Moldova), simultaneously, two years of studies, associate professor at the ”George Bacovia” University of Bacau; Head of the department “Programming Technologies”; professor, Moldova State University.
He has developed and implemented information systems for enterprises in the Food Industry of Moldova and the USSR, Chisinau City Hall, the Government of the Republic of Moldova, the Academy of Sciences of Moldova, the Higher Attestation Commission, the Ministry of Health, etc. He has over 100 publications, and has trained 10 doctors of science, under direct supervision or co-supervision.
Expert of the National Agency for Quality Assurance in Education and Research.
Expert of the Ministry of Education and Research of the Republic of Moldova. Expert of the Romanian Agency for Quality Assurance in Higher Education.”
Silvia Cirstea, Anglia Ruskin University
Contributions to the Development of Assistive Technologies
Abstract:
Deplasarea autonoma in medii nefamiliare, in particular in absenta GPS-ului sau in lumina slaba, este o problema importanta atat in tehnologiile asistive, cat si in robotica. Agentii autonomi sunt sisteme capabile sa se orienteze si sa se deplaseze in spatiu fara interventie externa. Ei utilizeaza o varietate de senzori (vizuali, acustici, inertiali) pentru a colecta informatii despre mediul inconjurator, pe care le proceseaza pentru a inainta pana la atingerea tintei. Exemple de tehnologii asistive portabile bazate pe agenti autonomi menite sa ajute persoanele cu deficiente de vedere sa se orienteze si sa se deplaseze independent vor fi prezentate. Acestea includ sisteme inteligente care exploateaza atat informatie vizuala, cat si informatie acustica.
Brief Bio:
Dr. Silvia Cirstea este conferentiar universitar si director-adjunct al Scolii Computing and Information Science la Anglia Ruskin University in Cambridge, UK, unde a initiat si dezvoltat cursuri de licenta si masterat in Inteligenta Artificiala si unde conduce din 2018 catedra de Computer Systems. Este licentiata a Facultatii de Matematica a Universitatii Bucuresti si are un doctorat in tehnologii imagistice obtinut la De Montfort University, Marea Britanie, care a inclus si un stagiu de cercetare la Princeton University, SUA. Dupa terminarea studiilor doctorale, a lucrat ca cercetator stiintific la Rutherford Appleton Laboratory (Central Laboratory of the UK Research Councils) si la Medical Research Council (UK), iar din 2005 este cadru didactic la Anglia Ruskin University. Silvia are peste 20 de ani de experienta in modelare matematica cu aplicatii in inginerie, fizica si stiinte medicale. Cercetarea ei se concentreaza pe procesarea de date si semnale, modelarea si optimizarea proceselor fizice si industriale (cu aplicatii diverse, e.g. in propagare electromagnetica, optica, imagistica medicala, acustica, sustenabilitate in constructii), statistica, probleme inverse si inteligenta artificiala. A lucrat in proiecte finantate de Uniunea Europeana, institutii publice din Marea Britanie (Innovate UK, UK Radio Communications Agency) si de parteneri industriali britanici (Effective Solutions Ltd, John Henry Group Ltd, TR Control Solutions). Silvia are un interes deosebit pentru autonomia asistata bazata pe informatie acustica pusa in slujba persoanelor cu deficiente de vedere, in particular in exploatarea potentialului oferit de inteligenta artificiala si de Internet-of-Things pentru a sprijini traiul independent al persoanelor cu astfel de deficiente. Este autoarea a numeroase articole stiintifice in publicatii prestigioase ca Scientific Reports, Journal of the Acoustical Society of America, Journal of the Optical Society of America, Experimental Brain Research or Frontiers, si este recenzor pentru organisme guvernamentale de finantare a cercetarii din Marea Britanie, Olanda, Italia si Croatia.