Memory-Based Pruning of Deep Neural Networks for IoT Devices Applied to Flood Detection.

Automatic flood detection may be an important component for triggering damage control systems and minimizing the risk of social or economic impacts caused by flooding. Riverside images from regular cameras are a widely available resource that can be used for tackling this problem. Nevertheless, state-of-the-art neural networks, the most suitable approach for this type of computer vision task, are usually resource-consuming, which poses a challenge for deploying these models within low-capability Internet of Things (IoT) devices with unstable internet connections. In this work, we propose a deep neural network (DNN) architecture pruning algorithm capable of finding a pruned version of a given DNN within a user-specified memory footprint. Our results demonstrate that our proposed algorithm can find a pruned DNN model with the specified memory footprint with little to no degradation of its segmentation performance. Finally, we show that our algorithm can be used in a memory-constraint wireless sensor network (WSN) employed to detect flooding events of urban rivers, and the resulting pruned models have competitive results compared with the original models.

The Relation of Satisfaction, Self-Confidence and Emotion in a Simulated Environment.

Clinical simulation allows discussions about improving the quality on the patient's care. This method have effectiveness on what concerns to satisfaction, self-confidence and student motivation. However, during the assessment, the students have emotional reactions that have tended to be overlooked. In view of this, this article seeks to identify and describe the relationship of the emotions observed by facial expressions and assess their degree of satisfaction and self-confidence by carrying out simulated practices among the nursing students. The analysis based on the scales showed high satisfaction and self-confidence levels, and it was found that the predominant basic emotion was anger, which is caused by other correlated emotions like tension and stress. This divergence between the identified emotions opens up space for further investigations about the level of motivation and the stimulus tolearning that these emotions can provide, and the extent to which they can lead to satisfaction and self-confidence.

How to Improve Fault Tolerance in Disaster Predictions: A Case Study about Flash Floods Using IoT, ML and Real Data.

The rise in the number and intensity of natural disasters is a serious problem that affects the whole world. The consequences of these disasters are significantly worse when they occur in urban districts because of the casualties and extent of the damage to goods and property that is caused. Until now feasible methods of dealing with this have included the use of wireless sensor networks (WSNs) for data collection and machine-learning (ML) techniques for forecasting natural disasters. However, there have recently been some promising new innovations in technology which have supplemented the task of monitoring the environment and carrying out the forecasting. One of these schemes involves adopting IP-based (Internet Protocol) sensor networks, by using emerging patterns for IoT. In light of this, in this study, an attempt has been made to set out and describe the results achieved by SENDI (System for dEtecting and forecasting Natural Disasters based on IoT). SENDI is a fault-tolerant system based on IoT, ML and WSN for the detection and forecasting of natural disasters and the issuing of alerts. The system was modeled by means of ns-3 and data collected by a real-world WSN installed in the town of São Carlos - Brazil, which carries out the data collection from rivers in the region. The fault-tolerance is embedded in the system by anticipating the risk of communication breakdowns and the destruction of the nodes during disasters. It operates by adding intelligence to the nodes to carry out the data distribution and forecasting, even in extreme situations. A case study is also included for flash flood forecasting and this makes use of the ns-3 SENDI model and data collected by WSN.

A Service-Oriented Middleware for Integrated Management of Crowdsourced and Sensor Data Streams in Disaster Management.

The increasing number of sensors used in diverse applications has provided a massive number of continuous, unbounded, rapid data and requires the management of distinct protocols, interfaces and intermittent connections. As traditional sensor networks are error-prone and difficult to maintain, the study highlights the emerging role of "citizens as sensors" as a complementary data source to increase public awareness. To this end, an interoperable, reusable middleware for managing spatial, temporal, and thematic data using Sensor Web Enablement initiative services and a processing engine was designed, implemented, and deployed. The study found that its approach provided effective sensor data-stream access, publication, and filtering in dynamic scenarios such as disaster management, as well as it enables batch and stream management integration. Also, an interoperability analytics testing of a flood citizen observatory highlighted even variable data such as those provided by the crowd can be integrated with sensor data stream. Our approach, thus, offers a mean to improve near-real-time applications.

NodePM: a remote monitoring alert system for energy consumption using probabilistic techniques.

In this paper, we propose an intelligent method, named the Novelty Detection Power Meter (NodePM), to detect novelties in electronic equipment monitored by a smart grid. Considering the entropy of each device monitored, which is calculated based on a Markov chain model, the proposed method identifies novelties through a machine learning algorithm. To this end, the NodePM is integrated into a platform for the remote monitoring of energy consumption, which consists of a wireless sensors network (WSN). It thus should be stressed that the experiments were conducted in real environments different from many related works, which are evaluated in simulated environments. In this sense, the results show that the NodePM reduces by 13.7% the power consumption of the equipment we monitored. In addition, the NodePM provides better efficiency to detect novelties when compared to an approach from the literature, surpassing it in different scenarios in all evaluations that were carried out.

A Mechanism for Verifying the Integrity and Immutability of Tuberculosis Data Using IOTA Distributed Ledger Technology.

BACKGROUND: Intensified research and innovation and rapid uptake of new tools, interventions, and strategies are crucial to fight Tuberculosis, the world's deadliest infectious disease. The sharing of health data remains a significant challenge. Data consumers must be able to verify the consistency and integrity of data. Solutions based on distributed ledger technologies may be adequate, where each member in a network holds a unique credential and stores an identical copy of the ledger and contributes to the collective process of validating and certifying digital transactions. OBJECTIVES: This work proposes a mechanism and presents a use case in Digital Health to allow the verification of integrity and immutability of TB electronic health records. METHODS: IOTA was selected as a supporting tool due to its data immutability, traceability and tamper-proof characteristics. RESULTS: A mechanism to verify the integrity of data through hash functions and the IOTA network is proposed. Then, a set of TB related information systems was integrated with the network. CONCLUSION: IOTA technology offers performance and flexibility to enable a reliable environment for electronic health records.

Using emotion recognition to assess simulation-based learning.

Simulation-based assessment relies on instruments that measure knowledge acquisition, satisfaction, confidence, and the motivation of students. However, the emotional aspects of assessment have not yet been fully explored in the literature. This dimension can provide a deeper understanding of the experience of learning in clinical simulations. In this study, a computer (software) model was employed to identify and classify emotions with the aim of assessing them, while creating a simulation scenario. A group of (twenty-four) students took part in a simulated nursing care scenario that included a patient suffering from ascites and respiratory distress syndrome followed by vomiting. The patient's facial expressions were recorded and then individually analyzed on the basis of six critical factors that were determined by the researchers in the simulation scenario: 1) student-patient communication, 2) dealing with the patient's complaint, 3) making a clinical assessment of the patient, 4) the vomiting episode, 5) nursing interventions, and 6) making a reassessment of the patient. The results showed that emotion recognition can be assessed by means of both dimensional (continuous models) and cognitive (discrete or categorical models) theories of emotion. With the aid of emotion recognition and classification through facial expressions, the researchers succeeded in analyzing the emotions of students during a simulated clinical learning activity. In the study, the participants mainly displayed a restricted affect during the simulation scenario, which involved negative feelings such as anger, fear, tension, and impatience, resulting from the difficulty of creating the scenario. This can help determine which areas the students were able to master and which caused them greater difficulty. The model employed for the recognition and analysis of facial expressions in this study is very comprehensive and paves the way for further use and a more detailed interpretation of its components.

Increasing Intelligence in Inter-Vehicle Communications to Reduce Traffic Congestions: Experiments in Urban and Highway Environments.

Intelligent Transportation Systems (ITS) rely on Inter-Vehicle Communication (IVC) to streamline the operation of vehicles by managing vehicle traffic, assisting drivers with safety and sharing information, as well as providing appropriate services for passengers. Traffic congestion is an urban mobility problem, which causes stress to drivers and economic losses. In this context, this work proposes a solution for the detection, dissemination and control of congested roads based on inter-vehicle communication, called INCIDEnT. The main goal of the proposed solution is to reduce the average trip time, CO emissions and fuel consumption by allowing motorists to avoid congested roads. The simulation results show that our proposed solution leads to short delays and a low overhead. Moreover, it is efficient with regard to the coverage of the event and the distance to which the information can be propagated. The findings of the investigation show that the proposed solution leads to (i) high hit rate in the classification of the level of congestion, (ii) a reduction in average trip time, (iii) a reduction in fuel consumption, and (iv) reduced CO emissions.