Dataset Information

Building test data from real outbreaks for evaluating detection algorithms.

ABSTRACT: Benchmarking surveillance systems requires realistic simulations of disease outbreaks. However, obtaining these data in sufficient quantity, with a realistic shape and covering a sufficient range of agents, size and duration, is known to be very difficult. The dataset of outbreak signals generated should reflect the likely distribution of authentic situations faced by the surveillance system, including very unlikely outbreak signals. We propose and evaluate a new approach based on the use of historical outbreak data to simulate tailored outbreak signals. The method relies on a homothetic transformation of the historical distribution followed by resampling processes (Binomial, Inverse Transform Sampling Method-ITSM, Metropolis-Hasting Random Walk, Metropolis-Hasting Independent, Gibbs Sampler, Hybrid Gibbs Sampler). We carried out an analysis to identify the most important input parameters for simulation quality and to evaluate performance for each of the resampling algorithms. Our analysis confirms the influence of the type of algorithm used and simulation parameters (i.e. days, number of cases, outbreak shape, overall scale factor) on the results. We show that, regardless of the outbreaks, algorithms and metrics chosen for the evaluation, simulation quality decreased with the increase in the number of days simulated and increased with the number of cases simulated. Simulating outbreaks with fewer cases than days of duration (i.e. overall scale factor less than 1) resulted in an important loss of information during the simulation. We found that Gibbs sampling with a shrinkage procedure provides a good balance between accuracy and data dependency. If dependency is of little importance, binomial and ITSM methods are accurate. Given the constraint of keeping the simulation within a range of plausible epidemiological curves faced by the surveillance system, our study confirms that our approach can be used to generate a large spectrum of outbreak signals.

SUBMITTER: Texier G

PROVIDER: S-EPMC5593515 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Building test data from real outbreaks for evaluating detection algorithms.

Texier Gaetan G Jackson Michael L ML Siwe Leonel L Meynard Jean-Baptiste JB Deparis Xavier X Chaudet Herve H

PloS one 20170901 9

Benchmarking surveillance systems requires realistic simulations of disease outbreaks. However, obtaining these data in sufficient quantity, with a realistic shape and covering a sufficient range of agents, size and duration, is known to be very difficult. The dataset of outbreak signals generated should reflect the likely distribution of authentic situations faced by the surveillance system, including very unlikely outbreak signals. We propose and evaluate a new approach based on the use of his ...[more]

PMID: 28863159

Similar Datasets

Project description:Well-powered genomic studies require genome-wide marker coverage across many individuals. For non-model species with few genomic resources, high-throughput sequencing (HTS) methods, such as Genotyping-By-Sequencing (GBS), offer an inexpensive alternative to array-based genotyping. Although affordable, datasets derived from HTS methods suffer from sequencing error, alignment errors, and missing data, all of which introduce noise and uncertainty to variant discovery and genotype calling. Under such circumstances, meaningful analysis of the data is difficult. Our primary interest lies in the issue of how one can accurately infer or impute missing genotypes in HTS-derived datasets. Many of the existing genotype imputation algorithms and software packages were primarily developed by and optimized for the human genetics community, a field where a complete and accurate reference genome has been constructed and SNP arrays have, in large part, been the common genotyping platform. We set out to answer two questions: 1) can we use existing imputation methods developed by the human genetics community to impute missing genotypes in datasets derived from non-human species and 2) are these methods, which were developed and optimized to impute ascertained variants, amenable for imputation of missing genotypes at HTS-derived variants? We selected Beagle v.4, a widely used algorithm within the human genetics community with reportedly high accuracy, to serve as our imputation contender. We performed a series of cross-validation experiments, using GBS data collected from the species Manihot esculenta by the Next Generation (NEXTGEN) Cassava Breeding Project. NEXTGEN currently imputes missing genotypes in their datasets using a LASSO-penalized, linear regression method (denoted 'glmnet'). We selected glmnet to serve as a benchmark imputation method for this reason. We obtained estimates of imputation accuracy by masking a subset of observed genotypes, imputing, and calculating the sample Pearson correlation between observed and imputed genotype dosages at the site and individual level; computation time served as a second metric for comparison. We then set out to examine factors affecting imputation accuracy, such as levels of missing data, read depth, minor allele frequency (MAF), and reference panel composition.

Project description:Despite extensive preventive efforts, falls continue to be a major source of morbidity and mortality among elderly. Real-time detection of falls and their urgent communication to a telecare center may enable rapid medical assistance, thus increasing the sense of security of the elderly and reducing some of the negative consequences of falls. Many different approaches have been explored to automatically detect a fall using inertial sensors. Although previously published algorithms report high sensitivity (SE) and high specificity (SP), they have usually been tested on simulated falls performed by healthy volunteers. We recently collected acceleration data during a number of real-world falls among a patient population with a high-fall-risk as part of the SensAction-AAL European project. The aim of the present study is to benchmark the performance of thirteen published fall-detection algorithms when they are applied to the database of 29 real-world falls. To the best of our knowledge, this is the first systematic comparison of fall detection algorithms tested on real-world falls. We found that the SP average of the thirteen algorithms, was (mean ± std) 83.0% ± 30.3% (maximum value = 98%). The SE was considerably lower (SE = 57.0% ± 27.3%, maximum value = 82.8%), much lower than the values obtained on simulated falls. The number of false alarms generated by the algorithms during 1-day monitoring of three representative fallers ranged from 3 to 85. The factors that affect the performance of the published algorithms, when they are applied to the real-world falls, are also discussed. These findings indicate the importance of testing fall-detection algorithms in real-life conditions in order to produce more effective automated alarm systems with higher acceptance. Further, the present results support the idea that a large, shared real-world fall database could, potentially, provide an enhanced understanding of the fall process and the information needed to design and evaluate a high-performance fall detector.

Project description:Linkage of medical databases, including insurer claims and electronic health records (EHRs), is increasingly common. However, few studies have investigated the behavior and output of linkage software. To determine how linkage quality is affected by different algorithms, blocking variables, methods for string matching and weight determination, and decision rules, we compared the performance of 4 nonproprietary linkage software packages linking patient identifiers from noninteroperable inpatient and outpatient EHRs. We linked datasets using first and last name, gender, and date of birth (DOB). We evaluated DOB and year of birth (YOB) as blocking variables and used exact and inexact matching methods. We compared the weights assigned to record pairs and evaluated how matching weights corresponded to a gold standard, medical record number. Deduplicated datasets contained 69,523 inpatient and 176,154 outpatient records, respectively. Linkage runs blocking on DOB produced weights ranging in number from 8 for exact matching to 64,273 for inexact matching. Linkage runs blocking on YOB produced 8 to 916,806 weights. Exact matching matched record pairs with identical test characteristics (sensitivity 90.48%, specificity 99.78%) for the highest ranked group, but algorithms differentially prioritized certain variables. Inexact matching behaved more variably, leading to dramatic differences in sensitivity (range 0.04-93.36%) and positive predictive value (PPV) (range 86.67-97.35%), even for the most highly ranked record pairs. Blocking on DOB led to higher PPV of highly ranked record pairs. An ensemble approach based on averaging scaled matching weights led to modestly improved accuracy. In summary, we found few differences in the rankings of record pairs with the highest matching weights across 4 linkage packages. Performance was more consistent for exact string matching than for inexact string matching. Most methods and software packages performed similarly when comparing matching accuracy with the gold standard. In some settings, an ensemble matching approach may outperform individual linkage algorithms.

Project description:BackgroundContact mixing plays a key role in the spread of COVID-19. Thus, mobility restrictions of varying degrees up to and including nationwide lockdowns have been implemented in over 200 countries. To appropriately target the timing, location, and severity of measures intended to encourage social distancing at a country level, it is essential to predict when and where outbreaks will occur, and how widespread they will be.MethodsWe analyze aggregated, anonymized health data and cell phone mobility data from Israel. We develop predictive models for daily new cases and the test positivity rate over the next 7 days for different geographic regions in Israel. We evaluate model goodness of fit using root mean squared error (RMSE). We use these predictions in a five-tier categorization scheme to predict the severity of COVID-19 in each region over the next week. We measure magnitude accuracy (MA), the extent to which the correct severity tier is predicted.ResultsModels using mobility data outperformed models that did not use mobility data, reducing RMSE by 17.3% when predicting new cases and by 10.2% when predicting the test positivity rate. The best set of predictors for new cases consisted of 1-day lag of past 7-day average new cases, along with a measure of internal movement within a region. The best set of predictors for the test positivity rate consisted of 3-days lag of past 7-day average test positivity rate, along with the same measure of internal movement. Using these predictors, RMSE was 4.812 cases per 100,000 people when predicting new cases and 0.79% when predicting the test positivity rate. MA in predicting new cases was 0.775, and accuracy of prediction to within one tier was 1.0. MA in predicting the test positivity rate was 0.820, and accuracy to within one tier was 0.998.ConclusionsUsing anonymized, macro-level data human mobility data along with health data aids predictions of when and where COVID-19 outbreaks are likely to occur. Our method provides a useful tool for government decision makers, particularly in the post-vaccination era, when focused interventions are needed to contain COVID-19 outbreaks while mitigating the collateral damage from more global restrictions.

Dataset Information

Building test data from real outbreaks for evaluating detection algorithms.

Publications

Building test data from real outbreaks for evaluating detection algorithms.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets