Real-time Surveillance System Research Articles

Marketing Little Cigars and Cigarillos: Advertising, Price, and Associations With Neighborhood Demographics

We have documented little cigar and cigarillo (LCC) availability, advertising, and price in the point-of-sale environment and examined associations with neighborhood demographics. We used a multimodal real-time surveillance system to survey LCCs in 750 licensed tobacco retail outlets that sold tobacco products in Washington, DC. Using multivariate models, we examined the odds of LCC availability, the number of storefront exterior advertisements, and the price per cigarillo for Black & Mild packs in relation to neighborhood demographics. The odds of LCC availability and price per cigarillo decreased significantly in nearly a dose-response manner with each quartile increase in proportion of African Americans. Prices were also lower in some young adult neighborhoods. Having a higher proportion of African American and young adult residents was associated with more exterior LCC advertising. Higher availability of LCCs in African American communities and lower prices and greater outdoor advertising in minority and young adult neighborhoods may establish environmental triggers to smoke among groups susceptible to initiation, addiction, and long-term negative health consequences.

STRAY DOG DETECTION IN WIRED CAMERA NETWORK

Existing surveillance systems impose high level of security on humans but lacks attention on animals. Stray dogs could be used as an alternative to humans to carry explosive material. It is therefore imperative to ensure the detection of stray dogs for necessary corrective action. In this paper, a novel composite approach to detect the presence of stray dogs is proposed. The captured frame from the surveillance camera is initially pre-processed using Gaussian filter to remove noise. The foreground object of interest is extracted utilizing ViBe algorithm. Histogram of Oriented Gradients (HOG) algorithm is used as the shape descriptor which derives the shape and size information of the extracted foreground object. Finally, stray dogs are classified from humans using a polynomial Support Vector Machine (SVM) of order 3. The proposed composite approach is simulated in MATLAB and OpenCV. Further it is validated with real time video feeds taken from an existing surveillance system. From the results obtained, it is found that a classification accuracy of about 96% is achieved. This encourages the utilization of the proposed composite algorithm in real time surveillance systems.

P-FAD: Real-Time Face Detection Scheme on Embedded Smart Cameras

Face detection on general embedded devices is fundamentally different from the conventional approach on personal computer or consumer digital camera due to the limited computation and power capacity. The resource-limited characteristic gives rise to new challenges for implementing a real-time video surveillance system with smart cameras. In this work, we present the design and implementation of Pyramid-like FAce Detection (P-FAD), a real-time face detection system constructed on general embedded devices. Motivated by the observation that the computation overhead increases proportionally to its pixel manipulation, P-FAD proposes a hierarchical approach to shift the complex computation to the promising regions. More specifically, P-FAD present a three-stage coarse, shift, and refine procedure, to construct a pyramid-like detection framework for reducing the computation overhead significantly. This framework also strikes a balance between the detection speed and accuracy. We have implemented P-FAD on notebook, Android phone and our embedded smart camera platform. An extensive system evaluation in terms of detailed experimental and simulation results is provided. Our empirical evaluation shows that P-FAD outperforms V-J detector calibrated color detector (VJ-CD) and color detector followed by a V-J detector (CD-VJ), the state of the art real-time face detection techniques by 4.7 -8.6 on notebook and by up to 8.2 on smart phone in terms of thembedded smart camera platfoe detection speed.

Detecting spatial-temporal clusters of HFMD from 2007 to 2011 in Shandong Province, China.

BackgroundHand, foot, and mouth disease (HFMD) has caused major public health concerns worldwide, and has become one of the leading causes of children death. China is the most serious epidemic area with a total of 3,419,149 reported cases just from 2008 to 2010, and its different geographic areas might have different spatial epidemiology characteristics at different spatial-temporal scale levels. We conducted spatial and spatial-temporal epidemiology analysis to HFMD at county level in Shandong Province, China.MethodsBased on the China National Disease Surveillance Reporting and Management System, the spatial-temporal database of HFMD from 2007 to 2011 was built. The global autocorrelation statistic (Moran’s I) was first used to detect the spatial autocorrelation of HFMD cases in each year. Purely Spatial scan statistics combined with Space-time scan statistic were used to detect epidemic clusters.ResultsThe annual average incidence rate was 93.70 per 100,000 in Shandong Province. Most HFMD cases (93.94%) were aged within 0–5 years old with an average male-to-female sex ratio 1.71, and the incidence seasonal peak was between April and July. The dominant pathogen was EV71 (47.35%), and CoxA16 (26.59%). HFMD had positive spatial autocorrelation at medium spatial scale level (county level) with higher Moran’s I from 0.31 to 0.62 (P<0.001). Seven spatial-temporal clusters were detected from 2007 to 2011 in the landscape of the whole Shandong, with EV71 or CoxA16 as the dominant pathogen for most hotspots areas.ConclusionsThe spatial-temporal clusters of HFMD wandered around the whole Shandong Province during 2007 to 2011, with EV71 or CoxA16 as the dominant pathogen. These findings suggested that a real-time spatial-temporal surveillance system should be established for identifying high incidence region and conducting prevention to HFMD timely.

A GIS-BASED ENVIRONMENTAL HEALTH INFORMATION SOURCE FOR MALAYSIAN CONTEXT

In environmental health studies, it is important to have good interaction between human and environment to produce a quality ecosystem. However, this is rather difficult to control due to mass economic development and natural disasters in the past few decades. These situations will eventually cause bad living environment to living species due to environmental hazards, exposures and health. In Malaysia, a real-time environmental health surveillance system has not been readily adopted. There are systems in European countries and the United States which may not be compatible for use in Malaysia. This research aims develop a real-time environmental health surveillance system to mitigate environmental health issues and assist public users to receive earlydetection emergency outbreaks near the location. A comprehensive environmental health ontological model was developed by integrating existing models of environmental health surveillance systems with customized map interfaces to assist public users to receive earlydetection of any emergency outbreaks near the location. The public users can also share and comment on the relevant environmental health outbreaks in social media networks. The proposed real-time environmental health management system (MyEHMS) has gone through the analysis, design, implementation and evaluation cycle. The testing and evaluation was conducted on the first prototype. The initial findings were used to model the final prototype and later the final prototype testing was tested on matching input against a dictionary of known patterns in database and public user query over the output accuracy and effectiveness. In MyEHMS accuracy testing, results showed that the proposed environmental health ontological model and the methods (N-gram approach and multi-tier approach) applied in environmental health management system was able to match most of the inputs from online media news and social media networks in both simulation platforms and real time platforms. However, constant revise of the environmental health ontological model is needed in order to continuously serve the increase number of environmental health issues. Besides, results in MyEHMS user testing also indicated that the environmental health management system has positive potential to be enhanced in order to assist public health in long run. Lastly, the overall results showed that the proposed environmental health ontological model adopted in MyEHMS is able to assist the public users to receive early-detection on emergency outbreaks near the location.

Real-time multi-camera video analytics system on GPU

In this article, parallel implementation of a real-time intelligent video surveillance system on Graphics Processing Unit (GPU) is described. The system is based on background subtraction and composed of motion detection, camera sabotage detection (moved camera, out-of-focus camera and covered camera detection), abandoned object detection, and object-tracking algorithms. As the algorithms have different characteristics, their GPU implementations have different speed-up rates. Test results show that when all the algorithms run concurrently, parallelization in GPU makes the system up to 21.88 times faster than the central processing unit counterpart, enabling real-time analysis of higher number of cameras.

Enabling Syndromic Surveillance in Pakistan

This work presents our first steps in developing a Global Real-time Infectious Disease Surveillance System (GRIDDS) employing robust and novel infectious disease epidemiology models with real-time inference and pre/exercise planning capabilities for Lahore, Pakistan. The objective of this work is to address the infectious disease surveillance challenges (specific to developing countries such as Pakistan) and develop a collaborative capability for monitoring and managing outbreaks of natural or manmade infectious diseases in Pakistan.

Influenza Forecasting with Google Flu Trends

We sought to develop a practical influenza forecast model, based on real-time, geographically focused, and easy to access data, to provide individual medical centers with advanced warning of the number of influenza cases, thus allowing sufficient time to implement an intervention. Secondly, we evaluated how the addition of a real-time influenza surveillance system, Google Flu Trends, would impact the forecasting capabilities of this model. The final model selection demonstrated that autoregression of influenza cases provides a strong base forecast model, which is enhanced by the addition of Google Flu Trends confirming the predictive capabilities of search query based syndromic surveillance.

Use of a Real-Time Syndromic Surveillance System to Improve Influenza Like Illness Screening and Documentation in Emergency Departments during the H1N1 Pandemic

Interventions introduced to increase ILI screening documentation exhibited a correlation with an improved documentation rate. Aggregated data across sites demonstrate that the greatest impact is associated with email reminders for recording ILI screening results, meetings on how to improve adherence and media broadcasts associated with the circulating pandemic influenza. When one site reliably reported a period of one-to-one nurse reminders to record the ILI screening result was analyzed, one of the strongest correlations to increased adherence was demonstrated. While the results suggest more direct interventions have a significant impact, further research to isolate which interventions had the greatest impact is warranted.

Influenza forecasting with Google Flu Trends.

BackgroundWe developed a practical influenza forecast model based on real-time, geographically focused, and easy to access data, designed to provide individual medical centers with advanced warning of the expected number of influenza cases, thus allowing for sufficient time to implement interventions. Secondly, we evaluated the effects of incorporating a real-time influenza surveillance system, Google Flu Trends, and meteorological and temporal information on forecast accuracy.MethodsForecast models designed to predict one week in advance were developed from weekly counts of confirmed influenza cases over seven seasons (2004–2011) divided into seven training and out-of-sample verification sets. Forecasting procedures using classical Box-Jenkins, generalized linear models (GLM), and generalized linear autoregressive moving average (GARMA) methods were employed to develop the final model and assess the relative contribution of external variables such as, Google Flu Trends, meteorological data, and temporal information.ResultsA GARMA(3,0) forecast model with Negative Binomial distribution integrating Google Flu Trends information provided the most accurate influenza case predictions. The model, on the average, predicts weekly influenza cases during 7 out-of-sample outbreaks within 7 cases for 83% of estimates. Google Flu Trend data was the only source of external information to provide statistically significant forecast improvements over the base model in four of the seven out-of-sample verification sets. Overall, the p-value of adding this external information to the model is 0.0005. The other exogenous variables did not yield a statistically significant improvement in any of the verification sets.ConclusionsInteger-valued autoregression of influenza cases provides a strong base forecast model, which is enhanced by the addition of Google Flu Trends confirming the predictive capabilities of search query based syndromic surveillance. This accessible and flexible forecast model can be used by individual medical centers to provide advanced warning of future influenza cases.

Generation of an index for physiological imbalance and its use as a predictor of primary disease in dairy cows during early lactation

Physiological imbalance (PI) is a situation in which physiological parameters deviate from the normal and cows consequently have an increased risk of developing production diseases and reduced production or reproduction. The objectives of this work were (1) to generate an index for PI based on several plasma metabolites and (2) to compare the use of this index with calculated energy balance (EBAL) and individual plasma metabolites in relation to risk of disease during early lactation. We used a total of 634 lactations from 317 cows consisting of 3 breeds ranging from a parity of 1 to 4. Weekly blood samples were analyzed for selected metabolites; that is, urea nitrogen, albumin, cholesterol, nonesterified fatty acids (NEFA), glucose, and β-hydroxybutyrate (BHBA). Energy intake and EBAL were calculated; veterinary treatment records and daily composite milk somatic cell counts were used to determine incidence of disease. Data were adjusted for numerous fixed effects (e.g., parity, breed, and week around calving) before further statistical analysis. The time of disease (TOD) was recorded as the day in which the signs of disease were observed (TOD=0). The week before and after TOD was ± n wk relative to TOD=0. Each week, all plasma metabolites were individually adjusted to an overall mean (=0) and variance (=1). The normalized variables were included in regression analyses by week of lactation to identify metabolites that explain the variation in calculated EBAL, as a reflection of degree of PI. Nonesterified fatty acids, BHBA, and glucose were weighted within each week based on regression coefficients (i.e., x1–x3 below) generated from a model to predict EBAL. Data from wk −1 relative to TOD were analyzed using a mixed linear model to relate degree of PI and metabolites in blood to risk of disease. The weekly PI index was defined as PI=(x1 × [NEFA])+x2 × [BHBA] − x3 × [glucose])/3. For diseases that developed ≥2 wk after calving, no variables were associated with risk of disease. Prepartal PI and plasma NEFA were better predictors of disease (i.e., metritis, retained placenta, and milk fever) at wk 1 than EBAL and plasma BHBA and glucose. Examining the relationship between PI and milk constituents is needed for the development of an automated in-line and real-time surveillance system for early detection of risk animals on-farm.

FGF-9 accelerates epithelial invagination for ectodermal organogenesis in real time bioengineered organ manipulation

BackgroundEpithelial invagination is important for initiation of ectodermal organogenesis. Although many factors regulate ectodermal organogenesis, there is not any report about their functions in real-time study. Electric cell-substrate impedance sensing (ECIS), a non-invasive, real-time surveillance system, had been used to detect changes in organ cell layer thickness through quantitative monitoring of the impedance of a cell-to-microelectrode interface over time. It was shown to be a good method for identifying significant real-time changes of cells. The purpose of this study is to establish a combined bioengineered organ-ECIS model for investigating the real time effects of fibroblast growth factor-9 (FGF-9) on epithelial invagination in bioengineered ectodermal organs. We dissected epithelial and mesenchymal cells from stage E14.5 murine molar tooth germs and identified the real-time effects of FGF-9 on epithelial-mesenchymal interactions using this combined bioengineered organ-ECIS model.ResultsMeasurement of bioengineered ectodermal organ thickness showed that Fibroblast growth factor-9 (FGF-9) accelerates epithelial invagination in reaggregated mesenchymal cell layer within 3 days. Gene expression analysis revealed that FGF-9 stimulates and sustains early Ameloblastin and Amelogenin expression during odontogenesis.ConclusionsThis is the first real-time study to show that, FGF-9 plays an important role in epithelial invagination and initiates ectodermal organogenesis. Based on these findings, we suggest FGF-9 can be applied for further study in ectodermal organ regeneration, and we also proposed that the ‘FGF-BMP balancing system’ is important for manipulating the morphogenesis of ectodermal organs. The combined bioengineered organ-ECIS model is a promising method for ectodermal organ engineering and regeneration research.

클러터환경에서 영상특징을 이용한 표적 추적

In this paper, we propose a novel tracking method which uses image features consisted of the area, average intensity, aspect ratio of a target image for the real-time IR surveillance system. The image features of the ground target can be modeled as a random process with exponential autocorrelation function mathematically. Finally, we derived a discrete target dynamic equation including kinematic states and geometric states of the target. Simulation results shows that the performance of the proposed method is better than that of the previous tracking method.

317 Drug-Related Problems in the Emergency Department: Development of a Real-Time Electronic Surveillance and Notification System

Study Objectives: We sought to develop an electronic notification system that uses readily available information from the electronic medical record to assist in the early identification of emergency department (ED) patients with drug-related problems. Methods: This prospective observational study was conducted in an urban academic ED from 12/9/11 to 1/10/12. Based on the published literature and investigator consensus, we selected 5 drug(s)/categories and related laboratory/clinical parameters that were likely to accurately identify ED patients with drug-related problems: 1) cardiovascular; 2) antidiabetic; 3) central nervous system; 4) antithrombotic; 5) specific drugs; and 6) combined triggers. We subsequently designed an electronic algorithm to send automated electronic notifications to investigators when any of these 6 criteria were electronically identified in the ED electronic medical record. Investigators, blinded to early notification system status, reviewed all patients to determine the incidence of drug-related problemss (gold standard). We calculated estimates of sensitivity, specificity, positive predictive value, negative predictive value and likelihood ratios by 2x2 contingency table analysis. Results: During the study period, there were 124 drug-related problemss in 514 (24.1%) patients reviewed. The median time from ED arrival to trigger of the early notification system was 94 minutes (IQR 87min). Mean age [56.6 yrs (SD: 20.9)] and sex (47.3% male) in the study cohort mirrored our ED population. The majority of drug-related problemss involved one (72%) or 2 (19%) medications. cardiovascular (23.1%), central nervous system (18.5%), endocrine (15.4%), anti-infectives (13.9%) and anti-neoplastic (12.3%) agents were associated with the majority of drug-related problemss. The sensitivity, specificity, positive predictive value, negative predictive value and likelihood ratio+ (95% CI) values for cardiovascular, antidiabetic and combined triggers categories were as follows: cardiovascular [50.0, 73.9, 37.8, 82.3, 1.9 (1.5-2.4)]; antidiabetic: [3.2, 99.2, 57.1, 76.3, 4.2 (1.01-16.5)]; combined triggers: [60.5, 68.2, 37.7, 84.4, 1.9 (1.5-2.3)]. Conclusion: The electronic algorithm and automated early notification system may be a useful tool for identification of drug-related problemss for some drug categories.

SprayQc: A Real-Time LC–MS/MS Quality Monitoring System To Maximize Uptime Using Off the Shelf Components

With the advent of high-throughput mass spectrometry (MS)-based proteomics, the magnitude and complexity of the performed experiments has increased dramatically. Likewise, investments in chromatographic and MS instrumentation are a large proportion of the budget of proteomics laboratories. Guarding measurement quality and maximizing uptime of the LC-MS/MS systems therefore requires constant care despite automated workflows. We describe a real-time surveillance system, called SprayQc, that continuously monitors the status of the peripheral equipment to ensure that operational parameters are within an acceptable range. SprayQc is composed of multiple plug-in software components that use computer vision to analyze electrospray conditions, monitor the chromatographic device for stable backpressure, interact with a column oven to control pressure by temperature, and ensure that the mass spectrometer is still acquiring data. Action is taken when a failure condition has been detected, such as stopping the column oven and the LC flow, as well as automatically notifying the appropriate operator. Additionally, all defined metrics can be recorded synchronized on retention time with the MS acquisition file, allowing for later inspection and providing valuable information for optimization. SprayQc has been extensively tested in our laboratory, supports third-party plug-in development, and is freely available for download from http://sourceforge.org/projects/sprayqc .

Nationwide Surveillance of Influenza during the Pandemic (2009–10) and Post-Pandemic (2010–11) Periods in Taiwan

IntroductionAlthough WHO declared the world moving into the post-pandemic period on August 10, 2010, influenza A(H1N1) 2009 virus continued to circulate globally. Its impact was expected to continue during the 2010–11 influenza season. This study describes the nationwide surveillance findings of the pandemic and post-pandemic influenza periods in Taiwan and assesses the impact of influenza A(H1N1) 2009 during the post-pandemic period.MethodsThe Influenza Laboratory Surveillance Network consisted of 12 contract laboratories for collecting and testing samples with acute respiratory tract infections. Surveillance of emergency room visits and outpatient department visits for influenza-like illness (ILI) were conducted using the Real-Time Outbreak and Disease Surveillance system and the National Health Insurance program data, respectively. Hospitalized cases with severe complications and deaths were reported to the National Notifiable Disease Surveillance System.ResultsDuring the 2009–10 influenza season, pandemic A(H1N1) 2009 was the predominant circulating strain and caused 44 deaths. However, the 2010–11 influenza season began with A(H3N2) being the predominant circulating strain, changing to A(H1N1) 2009 in December 2010. Emergency room and outpatient department ILI surveillance displayed similar trends. By March 31, 2011, there were 1,751 cases of influenza with severe complications; 50.1% reported underlying diseases. Of the reported cases, 128 deaths were associated with influenza. Among these, 93 (72.6%) were influenza A(H1N1) 2009 and 30 (23.4%) A(H3N2). Compared to the pandemic period, during the immediate post-pandemic period, increased number of hospitalizations and deaths were observed, and the patients were consistently older.ConclusionsReemergence of influenza A(H1N1) 2009 during the 2010–11 influenza season had an intense activity with age distribution shift. To further mitigate the impact of future influenza epidemics, Taiwan must continue its multifaceted influenza surveillance systems, remain flexible with antiviral use policies, and revise the vaccine policies to include the population most at risk.

Statistical skin color detection method without color transformation for real-time surveillance systems

Skin color is the significant information for many emerging applications in surveillance systems. However, the common skin color models usually need to perform color space transformation. This is not suitable for direct hardware implementation. This paper develops a statistical skin color model using the default RGB color space, which is especially suitable to implement on hardware for image processing applications. Moreover, an efficient face detection system is also proposed with our skin color model for hardware implementation. Compared with other skin color models, the proposed model produces the highest detection rate. Furthermore, the extended face detection system also significantly decreases the computational cost of the hardware implementation based on our skin color model. Experimental results demonstrate that our proposed detection system can be easily implemented on a field-programmable gate array (FPGA), where only 3202 logic cells is occupied with the high detection rate.

Using the National Poison Data System to Detect Mistaken Oral Ingestions of Medication Capsules Designed for Use in Pulmonary Inhalers

Introduction In 2006, the Institute of Medicine proposed initiatives to decrease medication errors using hospital monitoring systems for adverse event (AE) surveillance. Serendipitously, several poison center clinicians noticed a trend of inadvertent oral ingestions involving formoterol fumarate and tiotropium bromide. Each product is packaged in capsules for insertion into special pulmonary inhaler devices for chronic pulmonary disease. The National Poison Data System (NPDS) is a near real-time data and surveillance system capable of tracking AEs. Our objective was to assess the feasibility of NPDS to analyze outpatient AEs, employing formoterol and tiotropium ingestions as a model. Methods Retrospective review of NPDS data for formoterol and tiotropium mistaken oral ingestions (MOI) (2001–2007). Inclusion criteria were age over 6 years, exposures, all administration routes, all reasons, incorrect dosing route, and all medical outcomes. Primary outcome measures were number of MOI cases and demographics of at-risk patients. Results There were 3,919 formoterol and 18,096 tiotropium cases that met inclusion criteria. MOI cases with minor or moderate outcomes were formoterol 82 (2.2%) and tiotropium 131 (0.8%). Neither drug had a major medical outcome or death. Mean age and gender were similar for both medications. MOI was more common among female patients. Health care facility MOIs accounted for 19 (0.5%) formoterol and 35 (0.2%) tiotropium cases. From 2005 to 2007, formoterol and tiotropium cases accounted for 35% to 45% of all NPDS-tracked AEs. Conclusion NPDS data analysis revealed the magnitude of AEs for capsule-shaped medications for inhaler use was more common than previously identified.

Multidrug-resistant organisms in a community living facility: Tracking patient interactions and time spent in common areas

Contact precautions in community living facilities (CLF) are used to reduce the transmission of multidrug-resistant organisms (MDRO). However, this policy does not address the contamination of shared spaces, devices (eg, wheelchairs), and interactions with other patients. Using a real-time surveillance system, this study examines the time MDRO-positive patients spend interacting with others in communal areas. The findings from this study may be used to tailor MDRO policies and practices to the specific needs of CLF.

Speed Optimized Implementation of HUMMINGBIRD Cryptography for Sensor Network

The wireless sensor network (WSN) is well known for an enabling technology for the ubiquitous environment such as real-time surveillance system, habitat monitoring, home automation and healthcare applications. However, the WSN featuring wireless communication through air, a resource constraints device and irregular network topology, is threatened by malicious nodes such as eavesdropping, forgery, illegal modification or denial of services. For this reason, security in the WSN is key factor for utilizing the sensor network into the commercial way. There is a series of symmetric cryptography proposed by laboratory or industry for a long time. Among of them, recently proposed HUMMINGBIRD algorithm, motivated by the design of the well-known Enigma machine, is much more suitable to resource constrained devices, including smart card, sensor node and RFID tags in terms of computational complexity and block size. It also provides resistance to the most common attacks such as linear and differential cryptanalysis. In this paper, we implements ultra-lightweight cryptography, HUMMINGBIRD algorithm into the resource constrained device, sensor node as a perfectly customized design of sensor node.