Filippo Cavallo's Bio
Filippo Cavallo is Associate Professor in Biomedical Robotics at the University of Florence, Department of Industrial Engineering, Florence, Italy. He received the Master Degree in Electronics Engineering, Curriculum Bioengineering, from the University of Pisa, Italy, and the Ph.D. degree in Bioengineering at the BioRobotics Institute of Scuola Superiore Sant’Anna, Pisa, Italy. From 2007 to 2013, he was post doc researcher and, from 2013 to 2019, he was assistant professor and head and scientific responsible of the Assistive Robotics Lab at the BioRobotics Institute, Scuola Superiore Sant'Anna. Since 2020, he is associate professor with the University of Florence in Biomedical Robotics and Bio-Mechatronics and affiliate professor at the BioRobotics Institute, Scuola Superiore Sant'Anna. The objectives of his research activities are to promote and evaluate novel service robotics for active and healthy ageing, and to identify and validate disruptive healthcare paradigms for neurodegenerative and chronic diseases, focusing on prevention and support for physical and cognitive declines. The main scientific and technological challenges concern social robotics, human robot interaction, wearable sensors, Internet of Things and artificial intelligence for robot companion and healthcare applications. He has participated in various national and European projects and is the author of 180+ papers on conferences and ISI journals.
Social robot and IoT technologies in prevention, monitoring and rehabilitation: scenarios and challenges
Nowadays IoT and robotic technologies are experimenting an increasing number of opportunities to be exploited in several scenarios of daily living, from the home to the city and at work. This trend comes with several multidisciplinary scientific challenges, covering both technological fields and clinical areas related to neurodegenerative diseases, neuropsychology and various chronic diseases. Particularly, the integration of Robotics, Internet of Things and Artificial Intelligence is an interesting approach that enables the possibility to design and develop new frontiers in personalized and precision medicine, cognitive frailty and cooperative robotics. In this context, this talk aims to envisage possible scenarios and challenges of integrating BioRobotics and IoT technologies in healthcare and active and healthy ageing applications. The design and implementation of this technology addresses bioengineering and clinical aspects, from early diagnosis and objective assessment in diseases to rehabilitation, monitoring and therapeutic control and adaptation. Moreover, new developments and advancements in social robots with their interacting capabilities, wearable sensors, and artificial intelligence approaches plays an important role for a concrete deployment in real experimental setting.
Keywords: Social robotics, biomedical applications, Internet of Things, rehabilitation, monitoring
João Barreto's Bio
João P. Barreto holds a Ph.D. degree from the University of Coimbra (UC). He was a scholar in INRIA Rhone-Alpes and University of Pennsylvania, before joining the UC as a Professor. Joao is an expert in 3D Computer Vision, having authored more than 100 peer-reviewed articles, and regularly serving as Area Chair and Associate Editor in the most prestigious conferences and journals in the field. As an academic he received several distinctions and awards including the “Google Faculty Research Award”. Joao is also founder and CEO of Perceive3D SA (P3D), a spin-off company of UC that works in Computer Assisted Orthopeadic Surgery using Augmented Reality. As CEO he is responsible for R&D, Business Development and Investor Relations. P3D was recipient of the SMEi Phase 2 by the European Commission and the Bartolomeu de Gusmao award by the Portuguese Institute of Intellectual Property (INPI).
Navigation in Orthopeadic Surgery using Computer Vision and Augmented Reality
Arthroscopy is a modality of orthopeadic surgery where instruments and endoscopic camera (the arthroscope) are inserted into the articular cavity through small incisions (the surgical ports). Arthroscopy is highly beneficial for the patient and healthcare system because it reduces trauma, risk of infection and recovery time. However, clinical execution is difficult to accomplish because of indirect visualization and limited manoeuvrability inside the joint, with novices having to undergo a long training period and experts making mistakes of clinical consequences. This is a scenario where computer-aided surgery can have strong impact in improving clinical outcome and disseminating the benefits of minimally invasive surgery.
This talk introduces Video-based Surgical Navigation (VSN) that combines real-time video processing for accurate 3D measurements on the anatomy, with augmented reality to overlay guidance information in the images that is easily perceived by the surgeon. We will discuss the challenges in applying 3D computer vision to arthroscopic footage and the devised solutions that conducted to the first functional systems for navigation in Arthroscopy. We will also see how VSN can be extended to open orthopaedic surgery and will debate the path forward for augmented and mixed reality to become a standard in the OR broadly used by surgeons.
Keywords: 3D Computer Vision, Augmented Reality, Surgical Navigation
Amr Abdelkader's Bio
Amr Abdelkader spent thirty years of his career in the industry before joining the British University in Egypt. During this period, he served as an industrial automation consultant for many industrial processing sectors in Egypt and middle east. Currently Dr. Amr Abdelkader is a professor of dynamics and mechanical vibrations. Moreover he is the director of the Center of Excellence for Predictive Maintenance in the British University in Egypt. Dr. Abdelkader has over 25 years practical experience in teaching and providing consultancy services in predictive maintenance and machinery condition monitoring employing vibration testing and analysis. His main research interests are on machinery faults detection adopting Machine Learning and IOT.
Andreas Pester's Bio
Andreas Pester is a Mathematician and a Data Scientist. He is a Professor at the Faculty of Informatics and Computer Science at the British University in Egypt. He has more than 30 years of experience in teaching math and mathematical modelling, simulation technologies, online labs, and machine learning. He was previously responsible for the development of Data Science and Deep Learning research and teaching at CUAS (Austria) and BUE (Egypt). In addition to his extensive scientific publications, Professor Pester was involved in more than 20 EU- and national projects in Austria and Egypt. He is also a reviewer for Ph.D. promotions at UPC Barcelona, UNED Madrid and the University of Reading, UK. He was a Guest Professor at the UPC Barcelona, Technical University of Kharkov and Kiev, University Maribor, UNESP Bauru (Brazil), University of Applied Sciences Vienna, Armenian-American University in Yerevan, PUC Rio de Janeiro, and University Banja Luka.
Smart remote decentralized maintenance
As global population, water scarcity becomes an increasingly pressing challenge and water desalination becomes inevitable. The number of desalination plants is expected to grow, now there are more than 20,000 plant worldwide. More than 300 million people around the world rely on desalination for some or all their daily water needs. In Egypt, water is one of the governing elements for all national development plans.
The use of seawater and ground water desalination employing Reverse Osmosis (RO) technology offers a solution to Egypt’s water shortage challenge. In this process, a High-Pressure Pump (HPP) raises the pressure of the saline feed water. It consumes at least 50% of the plant’s electric energy. They are generally employed in small and medium RO desalination plants (5-1500 m3/d production capacity), located in remote area.
The current technical developments combine new HPPs and the rising demand for smart machines and services. Predictive Analytics, Remote Monitoring tools as well as Virtual Maintenance tools will be developed and integrated with new PD pumps. Such, smart features will allow for considerably reduced downtime resulting in cost savings, boost productivity, and facilitate operational decision-making in RO plants.
In the keynote the question, how machine and deep learning can be used for smart predictive maintenance of remotely located RO will occupy a wide space. Methods of machine learning are used to calibrate the parameters of the physical model, to fuse simulation, synthetic and experimental data and to provide a fault detection and classification in several fault classes of the PD pump. Also, the use of tiny machine learning to improve decentralized machine learning solutions will be tackled.
Keywords: Desalination, HPP pumps, smart and remote maintenance, machine learning, time series, fault classification, tiny ML