RESUMO
Undeniably, Internet of Things (IoT) devices are gradually getting better over time; and IoT-based systems play a significant role in our lives. The pervasiveness of the new essential service models is expanding, and includes self-driving cars, smart homes, smart cities, as well as promoting the development of some traditional fields such as agriculture, healthcare, and transportation; the development of IoT devices has not shown any sign of cooling down. On the one hand, several studies are coming up with many scenarios for IoT platforms, but some critical issues related to performance, speed, power consumption, availability, security, and scalability are not yet fully resolved. On the other hand, IoT devices are manufactured and developed by different organizations and individuals; hence, there is no unified standard (uniformity of IoT devices), i.e., sending and receiving messages among them and between them and the upper layer (e.g., edge devices). To address these issues, this paper proposes an IoT Platform called BMDD (Broker-less and Microservice architecture, Decentralized identity, and Dynamic transmission messages) that has a combination of two architectural models, including broker-less and microservices, with cutting-edge technologies such as decentralized identity and dynamic message transmission. The main contributions of this article are five-fold, including: (i) proposing broker-less and microservice for the IoT platform which can reduce single failure point of brokering architecture, easy to scale out and improve failover; (ii) providing a decentralized authentication mechanism which is suitable for IoT devices attribute (i.e., mobility, distributed); (iii) applying the Role-Based Access Control (RBAC) model for the authorization process; (iv) exploiting the gRPC protocol combined with the Kafka message queue enhances transmission rates, transmission reliability, and reduces power consumption in comparison with MQTT protocol; and (v) developing a dynamic message transmission mechanism that helps users communicate with any device, regardless of the manufacturer, since it provides very high homogeneity.
RESUMO
Coronavirus Disease 2019 (COVID-19) pandemic has been ferociously destroying global health and economics. According to World Health Organisation (WHO), until May 2021, more than one hundred million infected cases and 3.2 million deaths have been reported in over 200 countries. Unfortunately, the numbers are still on the rise. Therefore, scientists are making a significant effort in researching accurate, efficient diagnoses. Several studies advocating artificial intelligence proposed COVID diagnosis methods on lung images with high accuracy. Furthermore, some affected areas in the lung images can be detected accurately by segmentation methods. This work has considered state-of-the-art Convolutional Neural Network architectures, combined with the Unet family and Feature Pyramid Network (FPN) for COVID segmentation tasks on Computed Tomography (CT) scanner samples from the Italian Society of Medical and Interventional Radiology dataset. The experiments show that the decoder-based Unet family has reached the best (a mean Intersection Over Union (mIoU) of 0.9234, 0.9032 in dice score, and a recall of 0.9349) with a combination between SE ResNeXt and Unet++. The decoder with the Unet family obtained better COVID segmentation performance in comparison with Feature Pyramid Network. Furthermore, the proposed method outperforms recent segmentation state-of-the-art approaches such as the SegNet-based network, ADID-UNET, and A-SegNet + FTL. Therefore, it is expected to provide good segmentation visualizations of medical images.