Cloud Computing has emerged as a new concept for hosting and delivering services over Internet. It allows remote access to virtualized and shared computer hardware and/or software resources via the Internet to perform processing that are traditionally located on local servers and to pay only for the used resources. So the computing power and the storage necessary for an application are provided on demand using geographically distant servers hosted over internet. Cloud technology is made possible thanks to the significant innovations made during recent years, in particular the virtualization technology. We will present principles behind cloud computing, it’s essential characteristics, the different service models as well as the deployment models, the used technologies as well as the major service providers, ending with some new trends in cloud computing.
Security challenges within Industry 4.0
Abstract: Since its introduction in 2011, Industry 4.0 is gaining much more attention, from both individuals and organizations (companies, governments, universities ...). This latest revolution is mainly empowered by some emerging technologies such as IIoT, Big Data, XR, EC/FC/CC, 5G, 3D printing, smart-factories, smart-grid, smart- cities ...; and other evolving ones such as robotics, AI, and automation. Of course, doing business with this panoply of sophisticated technological trends requires continuous safety and deep security. In this sense, the use of cryptocurrencies will continue to spread and be more generalized, even regulated. Relevant security issues will appear with the massive deployment of cyber-physical systems along with their supporting technologies. The CIA triad of security objectives is still considered with others in this context, in fact that we require features like interoperability, customer accessibility, information transparency, M2M cooperation, decentralized decisions, AAA services. Our concern here is not to focus on the functioning of industrial processes, but it targets to shed light on the current and future security challenges that will unfortunately complexify the role of the security leaders and managers. These challenges can range from simple threats to more risky ones; Zero-hour attacks will happen tending to disturb the normal functioning or make damage of different components of the cyber-systems. Through this talk, we will cover most of these security issues, and discuss some countermeasures and recommended policies. Then, we will end up to what training is requiredto address these issues and give some well-known respected certifications of security (Security 4.0).
Pr. Noureddine KERZAZI
Professor at Mohammed V University (ENSIAS)
AI-assisted DR Diagnosis:Applications, Practitioner experiences, and Neuro-symbolic Techniques
Abstract: AI can enhance Diabetic Retinopathy (DR) diagnosis experiences for a diverse set of medical practitioners: from senior ophthalmologists who need help to speed up the process of diagnosis, automation, and data wrangling, all the way to students who seek hints when stuck with their learning process and first diagnoses. DR is one of the most chronic diseases, as a complication of long-term diabetes, that is the leading cause of vision loss in middle-aged people. However, the treatment of an early-stage detected disease is more effective than a disease detected at an advanced stage. Thorough this talk, I will present the challenges encountered within an ongoing project related to DR diagnosis with the service of ophthalmology at CHU Oujda. I will pass through two main approaches symbolic and Deep Learning and present preliminary results. Then, I will discuss the results and limitations. Finally, a few critical requirements in AI-assisted DR diagnosis are usability, precision, and trust; and they create opportunities for innovative medical experiences for both local and remote medicine. In this talk, I will explain these concepts. I will also describe several new opportunities in AI-assisted diagnosis, which can drive the next set of deep learning advances.