The COVID-19 crisis has developed, and not unjustifiably, a strong data aspect, requiring to handle, store and exchange large amounts of data in an efficient and secure way, not only among hospitals and clinics, but also between government services globally. In this project, we propose the development of an extendible prototype of a big data platform for COVID-19 data. We focus on the scalability and the security of this platform, by employing novel technologies, such as NoSQL databases, MapReduce analytics and Blockchain.
Conscious of AI, IoT, mobile and automotive products’ explosive growth and potential environmental impact, Dolphin Design is committed to accelerating the development of energy-efficient System-on-Chip for its customers. Dolphin Design continuously aims to meet the stringent power consumption targets for Artificial Intelligence of Things (AIoT) devices. The next generation of IoT devices will be wireless due to inconvenience, costs, or in some cases, the inability to wiring them. In addition, many of them might have strict size constraints.
The increasing complexity of large power distribution networks (PDN) has suggested that real-time monitoring systems (CMS) become crucial for both daily operation and emerging research. Future smart grids are expected to be assisted by but also operate for a vast number of sensor and actuator nodes. Such sensor/actuator nodes will demand a more consistent and reliable power supply in order to perform operation-monitoring tasks as well as collaborative edge computing and wireless communication functions such as 5G or even smart in-situ autonomous repairs of the power grids.
The Machine Learning team at Aifred Health is aiming to improve the performance of their core decision support models and the intern will assist in the development of a system that will intelligently iterate across dozens of unique model configurations to find the best performing one by leveraging previous configurations and their performance outcomes.
The result of this project (which will be demonstrated by a use case) can make health equipments to be used outside of hospitals. This is achieved by reducing the computation cost of running Deep Learning models by 3rd party tools and use our accelerator solution to run the size reduced and optimized model. This greatly helps to lower the barrier for using costly equipments and make them more affordable and reachable to people in need of these equipments.
ShipHaul Logistics aims at developing an Intelligent Freight Services platform towards the digitization, integration and optimization of freight management. The proposed project is the first step to provide the foundations and infrastructure and kickstart the development of the envisioned platform. To this end, we will emphasize evaluation, testing and comparative studies concerning the efficiency and capabilities of various software solutions for the constituent modules of the proposed platform.
We present a method for automatically estimating the lighting conditions from a single image. As opposed to most previous works which proposed methods that deal with individual aspects of the problem (e.g. indoors vs outdoors, parametric vs non-parametric), the proposed method unifies these ideas into a single, coherent framework. Our method will automatically estimate both parametric (individual light sources) and non-parametric (environment maps) lighting representations from both indoor and outdoor images.
The project will aim at collecting and studying 5G wireless network data latency to a) find ways to improve latency results (reduction, avoidance of peaks, etc.) and b) identify methods to optimize the distribution of network or application function within the network (i.e. find the optimal location of a function in a network ranging from the device, the edge, the core or the cloud).
Adaptive radiotherapy (ART) consists of adjusting irradiation at each treatment phase in response to changes in the patient’s body (such as weight loss) or from the patient’s change in position. Indeed, since initial dose plans are determined from standard CT, the initial dose distribution may vary and yield sub-optimal dose delivery. Intra-procedural imaging such as cone-beam CT (CBCT) can be used to adapt plans on a daily basis, but requires real-time performance with the patient on the table.