Real-time Alerts Research Articles

A real-time prototype of a water level monitor and wide area early flood warning

This article presents a real-time prototype of a water level monitor and wide area early flood warning. The prototype consists of two systems: the first is designed to provide real-time display and alarm alerts on a lab view program; the second is designed to measure water levels and display results via applications on a smartphone in real-time using the Blynk application, as well as to alert users of impending floods via the LINE application using Internet of Things (IoT) technology. The target user in this research study is Sado Phong Community, Khao Kho District, Phetchabun Province. This case study was designed for a stream with a water depth of 200 cm and a three-tiered alarm system: 140 cm, 150 cm and 170 cm of water height. The study results showed that when simulating each set level increase, the prototype could display real-time water level changes and give an alarm in both systems. The error value of the meter for the system displayed in the lab view was ± 1.11%. In comparison, the error value for the system displayed via an application on a smartphone for ultrasonic sensor 1 was ± 0.60% and ultrasonic sensor 2 was ± 0.80%, including the ability to notify a group of community members via the LINE application.

CycleGuard

Every year 41,000 cyclists die in road traffic-related incidents worldwide [47]. One of the most startling and infuriating conflicts that cyclists experience is the so-called "right hook". It refers to a vehicle striking a cyclist heading in the same direction by turning right into the cyclist. To prevent such a crash, this work presents CycleGuard, an acoustic-based collision detection system using smartphones. It is composed of a cheap commercial off-the-shelf (COTS) portable speaker that emits imperceptible high-frequency acoustic signals and a smartphone for reflected signal reception and analysis. Since received acoustic signals bear rich information of their reflecting objects, CycleGuard applies advanced acoustic ranging techniques to extract those information for traffic analysis. Cyclists are alerted if any pending right hook crashes are detected. Real-time alerts ensure that cyclists have sufficient time to react, apply brakes, and eventually avoid the hazard. To validate the efficacy of CycleGuard, we implement a proof-of-concept prototype and carry out extensive in-field experiments under a broad spectrum of settings. Results show that CycleGuard achieves up to 95% accuracy in preventing right hook crashes and is robust to various scenarios. It is also energy-friendly to run on battery-powered smartphones.

A New Prototype of Smart Wearable Monitoring System Solution for Alzheimer's Patients.

PurposeThe daily life management of patients with Alzheimer’s disease (AD) constitutes a significant and rapidly expanding health-care responsibility. In this study, an innovative prototype of a wireless-sensing smart wearable medical device (SWMD) is proposed as a multi-functions solution for Alzheimer patients. The SWMD is aimed to assemble three main biomedical engineering advances: 1) use of a Wi-Fi microcontroller, 2) simultaneous monitoring of a set of vital biomarkers, and 3) cautions of fall down conditions, in addition to GPS location indicator.MethodsThe SWMD employs a Wi-Fi controller that is incorporated with electronic circuits to monitor three vital signals (temperature, heart rate, and oxygen saturation), fall down conditions in three directions (X, Y, and Z axis), and GPS location. The SWMD was connected to the Firebase Service (database hosted on the Internet Cloud). The proposed device was tested on 13 normal volunteers. The left side, right side, forward, and backward fall down conditions were assessed. The prototype’s functions during daily activity such as rising hand, sitting down or standing up, and walking conditions were also assessed.ResultsThe three assembled functions were all successfully incorporated to build the SWDM device as a suggested solution offering real-time alerts during daily activity to AD patients. The Bland-Altman statistical test showed no significant difference (p-value >0.05) between the SWMD biomarkers’ acquisition and the reference methods. The gyro/accelerator sensor yielded 93% sensitivity in fall down detection and 95% specificity during daily activities. The GPS yielded correct positioning of the SWDM holder, while the internet cloud allowed saving and managing all vital biomarkers daily.ConclusionThe SWMD is a possible solution for daily life support for AD patients. It incorporates three functions in one single device, GPS location indicator, monitoring set of biomarkers, and fall down alert, which are all controlled via a Wi-Fi micro controller on-line connected to Internet Cloud. It successfully would allow the management of the daily records as well as the real-time alerts to remote persons.

Implementation of IceTop data in the IceCube Realtime Alert System

The IceCube Neutrino Observatory is a cubic-kilometer detector at the geographic South Pole searching for astrophysical neutrinos. The main background for astrophysical neutrinos is leptons produced by cosmic ray air showers: neutrinos from the northern hemisphere and muons from the southern hemisphere. These atmospheric backgrounds are reduced using a data-driven selection based on the observed event energy and arrival direction. An array of ice-Cherenkov tanks on the surface, IceTop, detects showers of secondary particles created in cosmic ray air showers. We will show that data from IceTop can be used to further reduce the background due to atmospheric muons, and by implementing this method, we vetoed three alert events.

IoT-Fog-Cloud Centric Earthquake Monitoring and Prediction

Earthquakes are among the most inevitable natural catastrophes. The uncertainty about the severity of the earthquake has a profound effect on the burden of disaster and causes massive economic and societal losses. Although unpredictable, it can be expected to ameliorate damage and fatalities, such as monitoring and predicting earthquakes using the Internet of Things (IoT). With the resurgence of the IoT, an emerging innovative approach is to integrate IoT technology with Fog and Cloud Computing to augment the effectiveness and accuracy of earthquake monitoring and prediction. In this study, the integrated IoT-Fog-Cloud layered framework is proposed to predict earthquakes using seismic signal information. The proposed model is composed of three layers: (i) at sensor layer, seismic data are acquired, (ii) fog layer incorporates pre-processing, feature extraction using fast Walsh–Hadamard transform (FWHT), selection of relevant features by applying High Order Spectral Analysis (HOSA) to FWHT coefficients, and seismic event classification by K-means accompanied by real-time alert generation, (iii) at cloud layer, an artificial neural network (ANN) is employed to forecast the magnitude of an earthquake. For performance evaluation, K-means classification algorithm is collated with other well-known classification algorithms from the perspective of accuracy and execution duration. Implementation statistics indicate that with chosen HOS features, we have been able to attain high accuracy, precision, specificity, and sensitivity values of 93.30%, 96.65%, 90.54%, and 92.75%, respectively. In addition, the ANN provides an average correct magnitude prediction of 75%. The findings ensured that the proposed framework has the potency to classify seismic signals and predict earthquakes and could therefore further enhance the detection of seismic activities. Moreover, the generation of real-time alerts further amplifies the effectiveness of the proposed model and makes it more real-time compatible.

Clinical Utility and Physician Perceptions of a Digital Platform for Electronic Patient-Reported Outcomes Monitoring in Patients with Hematologic Malignancies in Real-World Practice

Clinical Utility and Physician Perceptions of a Digital Platform for Electronic Patient-Reported Outcomes Monitoring in Patients with Hematologic Malignancies in Real-World Practice

Using Surveillance With Near-Real-Time Alerts During a Cluster of Overdoses From Fentanyl-Contaminated Crack Cocaine, Connecticut, June 2019.

In 2019, Connecticut launched an opioid overdose-monitoring program to provide rapid intervention and limit opioid overdose-related harms. The Connecticut Statewide Opioid Response Directive (SWORD)-a collaboration among the Connecticut State Department of Public Health, Connecticut Poison Control Center (CPCC), emergency medical services (EMS), New England High Intensity Drug Trafficking Area (HIDTA), and local harm reduction groups-required EMS providers to call in all suspected opioid overdoses to the CPCC. A centralized data collection system and the HIDTA overdose mapping tool were used to identify outbreaks and direct interventions. We describe the successful identification of a cluster of fentanyl-contaminated crack cocaine overdoses leading to a rapid public health response. On June 1, 2019, paramedics called in to the CPCC 2 people with suspected opioid overdose who reported exclusive use of crack cocaine after being resuscitated with naloxone. When CPCC specialists in poison information followed up on the patients' status with the emergency department, they learned of 2 similar cases, raising suspicion that a batch of crack cocaine was mixed with an opioid, possibly fentanyl. The overdose mapping tool pinpointed the overdose nexus to a neighborhood in Hartford, Connecticut; the CPCC supervisor alerted the Connecticut State Department of Public Health, which in turn notified local health departments, public safety officials, and harm reduction groups. Harm reduction groups distributed fentanyl test strips and naloxone to crack cocaine users and warned them of the dangers of using alone. The outbreak lasted 5 days and tallied at least 22 overdoses, including 6 deaths. SWORD's near-real-time EMS reporting combined with the overdose mapping tool enabled rapid recognition of this overdose cluster, and the public health response likely prevented additional overdoses and loss of life.

Development, Validation, and Assessment of Clinical Impact of Real-time Alerts to Detect Inpatient As-Needed Opioid Orders With Duplicate Indications: Prospective Study.

BackgroundAs-needed (PRN) opioid orders with duplicate indications can lead to medication errors and opioid-related adverse drug events.ObjectiveThe objective of our study was to build and validate real-time alerts that detect duplicate PRN opioid orders and assist clinicians in optimizing the safety of opioid orders.MethodsThis single-center, prospective study used an iterative, 3-step process to refine alert performance by advancing from small sample evaluations of positive predictive values (PPVs) (step 1) through intensive evaluations of accuracy (step 2) to evaluations of clinical impact (step 3). Validation cohorts were randomly sampled from eligible patients for each step.ResultsDuring step 1, the PPV was 100% (one-sided, 97.5% CI 70%-100%) for moderate and severe pain alerts. During step 2, duplication of 1 or more PRN opioid orders was identified for 17% (34/201; 95% CI, 12%-23%) of patients during chart review. This bundle of alerts showed 94% sensitivity (95% CI 80%-99%) and 96% specificity (95% CI 92%-98%) for identifying patients who had duplicate PRN opioid orders. During step 3, at least 1 intervention was made to the medication profile for 77% (46/60; 95% CI 64%-87%) of patients, and at least 1 inappropriate duplicate PRN opioid order was discontinued for 53% (32/60; 95% CI 40%-66%) of patients.ConclusionsThe bundle of alerts developed in this study was validated against chart review by a pharmacist and identified patients who benefited from medication safety interventions to optimize PRN opioid orders.

High-Energy Alerts in the Multi-Messenger Era

The observation of electromagnetic counterparts to both high energy neutrinos and gravitational waves marked the beginning of a new era in astrophysics. The multi-messenger approach allows us to gain new insights into the most energetic events in the Universe such as gamma-ray bursts, supernovas, and black hole mergers. Real-time multi-messenger alerts are the key component of the observational strategies to unravel the transient signals expected from astrophysical sources. Focusing on the high-energy regime, we present a historical perspective of multi-messenger observations, the detectors and observational techniques used to study them, the status of the multi-messenger alerts and the most significant results, together with an overview of the future prospects in the field.

A smart mask for active defense against airborne pathogens

Face masks are a primary preventive measure against airborne pathogens. Thus, they have become one of the keys to controlling the spread of the COVID-19 virus. Common examples, including N95 masks, surgical masks, and face coverings, are passive devices that minimize the spread of suspended pathogens by inserting an aerosol-filtering barrier between the user’s nasal and oral cavities and the environment. However, the filtering process does not adapt to changing pathogen levels or other environmental factors, which reduces its effectiveness in real-world scenarios. This paper addresses the limitations of passive masks by proposing ADAPT, a smart IoT-enabled “active mask”. This wearable device contains a real-time closed-loop control system that senses airborne particles of different sizes near the mask by using an on-board particulate matter (PM) sensor. It then intelligently mitigates the threat by using mist spray, generated by a piezoelectric actuator, to load nearby aerosol particles such that they rapidly fall to the ground. The system is controlled by an on-board micro-controller unit that collects sensor data, analyzes it, and activates the mist generator as necessary. A custom smartphone application enables the user to remotely control the device and also receive real-time alerts related to recharging, refilling, and/or decontamination of the mask before reuse. Experimental results on a working prototype confirm that aerosol clouds rapidly fall to the ground when the mask is activated, thus significantly reducing PM counts near the user. Also, usage of the mask significantly increases local relative humidity levels.

Open Issues in Statistical Forecasting of Solar Proton Events: A Machine Learning Perspective

AbstractSeveral techniques have been developed in the last two decades to forecast the occurrence of Solar Proton Events (SPEs), mainly based on the statistical association between the 10 MeV proton flux and precursor parameters. The Empirical model for Solar Proton Events Real Time Alert (ESPERTA; Laurenza et al., 2009, https://doi.org/10.1029/2007sw000379) provides a quite good and timely prediction of SPEs after the occurrence of M2 soft x‐ray (SXR) bursts, by using as input parameters the flare heliolongitude, the SXR and the 1 MHz radio fluence. Here, we reinterpret the ESPERTA model in the framework of machine learning and perform a cross validation, leading to a comparable performance. Moreover, we find that, by applying a cut‐off on the M2 flares heliolongitude, the False Alarm Rate (FAR) is reduced. The cut‐off is set to where the cumulative distribution of M2 flares associated with SPEs shows a break which reflects the poor magnetic connection between the Earth and eastern hemisphere flares. The best performance is obtained by using the SMOTE algorithm, leading to probability of detection of 0.83 and a FAR of 0.39. Nevertheless, we demonstrate that a relevant FAR on the predictions is a natural consequence of the sample base rates. From a Bayesian point of view, we find that the FAR explicitly contains the prior knowledge about the class distributions. This is a critical issue of any statistical approach, which requires to perform the model validation by preserving the class distributions within the training and test datasets.

Automation in Surveillance: A Review of the Developments

Abstract: The purpose of this paper is to conduct a literature survey on the developments of surveillance systems while laying strong emphasis on AI-related developments. With the rise in availability of video data and advancements in computer vision techniques, the role of AI in surveillance has become more prominent. Most surveillance footage is used for investigative purposes after an event has occurred instead of providing real-time alerts. The systems discussed in this paper help achieve this goal of providing real-time alerts. Limitations and strengths of the systems are also discussed.

Synergistic Integration of Chemo-Resistive and SERS Sensing for Label-Free Multiplex Gas Detection.

Practical sensing applications such as real-time safety alerts and clinical diagnoses require sensor devices to differentiate between various target molecules with high sensitivity and selectivity, yet conventional devices such as oxide-based chemo-resistive sensors and metal-based surface-enhanced Raman spectroscopy (SERS) sensors usually do not satisfy such requirements. Here, a label-free, chemo-resistive/SERS multimodal sensor based on a systematically assembled 3D cross-point multifunctional nanoarchitecture (3D-CMA), which has unusually strong enhancements in both "chemo-resistive" and "SERS" sensing characteristics is introduced. 3D-CMA combines several sensing mechanisms and sensing elements via 3D integration of semiconducting SnO2 nanowire frameworks and dual-functioning Au metallic nanoparticles. It is shown that the multimodal sensor can successfully estimate mixed-gas compositions selectively and quantitatively at the sub-100 ppm level, even for mixtures of gaseous aromatic compounds (nitrobenzene and toluene) with very similar molecular structures. This is enabled by combined chemo-resistive and SERS multimodal sensing providing complementary information.

DriverSonar

Dangerous driving due to drowsiness and distraction is the main cause of traffic accidents, resulting in casualties and economic loss. There is an urgent need to address this problem by accurately detecting dangerous driving behaviors and generating real-time alerts. Inspired by the observation that dangerous driving actions induce unique acoustic features that respond to the signal of an acoustic source, we present the DriverSonar system in this paper. The proposed system detects dangerous driving actions and generates real-time alarms using off-the-shelf smartphones. Compared with the state-of-the-arts, the DriverSonar system does not require dedicated sensors but just uses the built-in speaker and microphone in a smartphone. Specifically, DriverSonar is able to recognize head/hand motions such as nodding, yawning, and abrupt adjustment of the steering wheel. We design, implement and evaluate DriverSonar with extensive experiments. We conduct both simulator-based and and real driving-based experiments (IRB-approved) with 30 volunteers for a period over 12 months. Experiment results show that the proposed system can detect drowsy and distraction related dangerous driving actions at an precision up to 93.2% and a low false acceptance rate of 3.6%.

Design Of Detection System With Motion Sensor And Notification On Android Based On Arduino Microcontroller

The rapid development of technology nowadays makes people want to always be creative and triggers to create something that is needed to create a security system, because of today's economy makes many people act criminally by robbery or theft. An alternative security system that is often used today is to use CCTV (Closed Circuit Television), but the limitations in conducting CCTV control include that it still has to be monitored directly at the monitoring location only. From the problems that occur, the authors develop a thesis that is able to control and monitor by sending realtime alerts in the form of notifications to the user on an Android Smartphone device by building an IoT (Internet of Thing) based detection system using a PIR (Passive Infrared) motion sensor entitled "Design And Build A Detection System With Motion Sensor And Notification On Android Based On Arduino Microcontroller".

Prediction of Crime Scene Objects Using Deep Learning

The crime rates today can inevitably put a civilian's life in danger. While consistent efforts are being made to alleviate crime, there is also a dire need to create a smart and proactive surveillance system. Our project implements a smart surveillance system that would alert the authorities in real-time when a crime is being committed. During armed robberies and hostage situations, most often, the police cannot reach the place on time to prevent it from happening, owing to the lag in communication between the informants of the crime scene and the police. We propose an object detection model that implements deep learning algorithms to detect objects of violence such as pistols, knives, rifles from video surveillance footage, and in turn send real-time alerts to the authorities. There are a number of object detection algorithms being developed, each being evaluated under the performance metric mAP. On implementing Faster R-CNN with ResNet 101 architecture we found the mAP score to be about 91%. However, the downside to this is the excessive training and inferencing time it incurs. On the other hand, YOLOv5 architecture resulted in a model that performed very well in terms of speed. Its training speed was found to be 0.012 s / image during training but naturally, the accuracy was not as high as Faster R-CNN. With good computer architecture, it can run at about 40 fps. Thus, there is a tradeoff between speed and accuracy and it's important to strike a balance. We use transfer learning to improve accuracy by training the model on our custom dataset. This project can be deployed on any generic CCTV camera by setting up a live RTSP (real-time streaming protocol) and streaming the footage on a laptop or desktop where the deep learning model is being run.

Improving Blood Pressure Accuracy in the Outpatient Adolescent Setting

Hypertension (HTN) is increasing in the pediatric population, and hypertensive children become hypertensive adults. Blood pressure (BP) is often overlooked due to factors including distrust in readings, lack of time in visits, and discomfort prescribing antihypertensive medications. The objective of our multistage, clinically based quality improvement project was to improve BP measurement and HTN diagnosis and intervention in adolescents. Study investigators performed interventions in an adolescent medicine clinic at Children's Hospital Colorado (ages 12-24 years), which included equipment inventory, mapping clinic processes, manual/automated BP training for staff, education of faculty/staff on guidelines, and creation/implementation of updated best practice alerts (BPAs) based on age-appropriate guidelines for stage 1, 2 HTN in patients younger/older 18years. With equipment updates, medical assistant manual BP certifications, educational sessions for faculty/staff, and creation of a BPA with instruction before the go-live date, confidence in using automated and manual BP measurements increased for faculty/staff. The number of unique patient visits presenting with elevated BPs decreased significantly in the postintervention period reflecting the education/training. Staff used the new order set minimally, but there was an increase in correct diagnoses of elevated BP and laboratory workup. Having appropriate equipment while introducing BP guideline education in the ambulatory setting with electronic health record reminders and utility of BPAs can decrease erroneous BP values saving providers and staff encounter time. Real-time alerts can aid in accurate diagnosis rates and improved intervention for youth with elevated BP readings. Providers still inconsistently interact with order sets despite such parameters.

Smartphone application alerts for early trauma team activation: Millennial technology in healthcare

BackgroundData access through smartphone applications (apps) has reframed procedure and policy in healthcare, but its impact in trauma remains unclear. Citizen is a free app that provides real-time alerts curated from 911 dispatch data. Our primary objective was to determine whether app alerts occurred earlier than recorded times for trauma team activation and emergency department arrival. MethodsTrauma registry entries were extracted from a level one urban trauma center from January 1, 2018 to June 30, 2019 and compared with app metadata from the center catchment area. We matched entries to metadata according to description, date, time, and location then compared metadata timestamps to trauma team activation and emergency department arrival times. We computed percentage of time the app reported traumatic events earlier than trauma team activation or emergency department arrival along with exact binomial 95% confidence interval; median differences between times were presented along with interquartile ranges. ResultsOf 3,684 trauma registry entries, 209 (5.7%) matched app metadata. App alerts were earlier for 96.1% and 96.2% of trauma team activation and emergency department arrival times, respectively, with events reported median 36 (24–53, IQR) minutes earlier than trauma team activation and 32 (25–42, IQR) minutes earlier than emergency department arrival. Registry entries for younger males, motor vehicle-related injuries and penetrating traumas were more likely to match alerts (P < .0001). ConclusionApps like Citizen may provide earlier notification of traumatic events and therefore earlier mobilization of trauma service resources. Earlier notification may translate into improved patient outcomes. Additional studies into the benefit of apps for trauma care are warranted.

Light Monitoring System using Z-Score Analysis

In today’s modern world where everything is being automated and security is a growing concern, we made an automated module to live-monitor the anomalies in any provided space at all times to ensure security in our personal space. By implementing our project, we can monitor anything important which would be out of our reach at the moment with a live alert system through which we can identify any anomalies. In our proposed system we integrated Machine Learning to work with an IoT system by using Bolt Wi-Fi module which also uses an LDR sensor to detect the light intensity, here LDR is used specifically to better understand the Z-Score analysis. We are using ML to do an analysis known as Z-Score, which processes a math equation to detect anomalies. This analysis is done to predict a frame of upper and lower boundaries for the light intensity. Eventually, when the LDR sensor value i.e., light intensity goes out of range in a room, it generates Real-Time alerts in the form of an SMS alert which will be directed to the user's mobile phone through Twilio. This alert system is an advanced way to increase the work efficiency of any live monitoring system as the ML is always working to increase accuracy. In our project, this system specifically uses Light Dependent Resistor to detect changes in light intensities, but this can be implemented for any sensor to detect.

Design and implementation of IoT based environment pollution monitoring system: A case study of Iraq

Researches confirm the existence of widespread pollution in Iraq, where a large number of citizens suffer from symptoms related to environmental pollutants. This is accompanied by the lack of an integrated monitoring system that provides wide-scale, real-time monitoring of air pollution for pollutants (Particulate Matter (PM), Sulfur Dioxide (SO2), Ground-Level Ozone (O3), Carbon Monoxide (CO), and Nitrogen Dioxide (NO2), ionizing radiation). In this paper, an integrated system was designed and implemented to meet the requirements of air pollution monitoring in Iraq on a wide-scale and provide its results in real-time to the end-users in community service as a significant step towards smart cities. Distributed computing concepts and design patterns were used to realize the system in the form of web services along with the embedded sensing units. Where the readings from the former are sent in real-time to the later to be aggregated and analysed in database systems. Reporting services are run on the data in web-services to inform the citizens with the air quality, radioactively polluted areas, and provide real-time alerts and recommendations to those with respiratory diseases.