Outlier Vehicle Trajectory Detection Using Deep Autoencoders in Santiago, Chile.

In the last decade, a large amount of data from vehicle location sensors has been generated due to the massification of GPS systems to track them. This is because these sensors usually include multiple variables such as position, speed, angular position of the vehicle, etc., and, furthermore, they are also usually recorded in very short time intervals. On the other hand, routes are often generated so that they do not correspond to reality, due to artifacts such as buildings, bridges, or sensor failures and where, due to the large amount of data, visual analysis of human expert is unable to detect genuinely anomalous routes. The presence of such abnormalities can lead to faulty sensors being detected which may allow sensor replacement to reliably track the vehicle. However, given the reliability of the available sensors, there are very few examples of such anomalies, which can make it difficult to apply supervised learning techniques. In this work we propose the use of unsupervised deep neural network models based on stacked autoencoders to detect anomalous routes in vehicles within Santiago de Chile. The results show that the proposed model is capable of effectively detecting anomalous paths in real data considering validation given by an expert user, reaching a performance of 82.1% on average. As future work, we propose to incorporate the use of Long Short-Term Memory (LSTM) and attention-based networks in order to improve the detection of anomalous trajectories.

Prior and Posterior Linear Pooling for Combining Expert Opinions: Uses and Impact on Bayesian Networks-The Case of the Wayfinding Model.

The use of expert knowledge to quantify a Bayesian Network (BN) is necessary when data is not available. This however raises questions regarding how opinions from multiple experts can be used in a BN. Linear pooling is a popular method for combining probability assessments from multiple experts. In particular, Prior Linear Pooling (PrLP), which pools opinions and then places them into the BN, is a common method. This paper considers this approach and an alternative pooling method, Posterior Linear Pooling (PoLP). The PoLP method constructs a BN for each expert, and then pools the resulting probabilities at the nodes of interest. The advantages and disadvantages of these two methods are identified and compared and the methods are applied to an existing BN, the Wayfinding Bayesian Network Model, to investigate the behavior of different groups of people and how these different methods may be able to capture such differences. The paper focusses on six nodes Human Factors, Environmental Factors, Wayfinding, Communication, Visual Elements of Communication and Navigation Pathway, and three subgroups Gender (Female, Male), Travel Experience (Experienced, Inexperienced), and Travel Purpose (Business, Personal), and finds that different behaviors can indeed be captured by the different methods.

Importance of different types of prior knowledge in selecting genome-wide findings for follow-up.

Biological plausibility and other prior information could help select genome-wide association (GWA) findings for further follow-up, but there is no consensus on which types of knowledge should be considered or how to weight them. We used experts' opinions and empirical evidence to estimate the relative importance of 15 types of information at the single-nucleotide polymorphism (SNP) and gene levels. Opinions were elicited from 10 experts using a two-round Delphi survey. Empirical evidence was obtained by comparing the frequency of each type of characteristic in SNPs established as being associated with seven disease traits through GWA meta-analysis and independent replication, with the corresponding frequency in a randomly selected set of SNPs. SNP and gene characteristics were retrieved using a specially developed bioinformatics tool. Both the expert and the empirical evidence rated previous association in a meta-analysis or more than one study as conferring the highest relative probability of true association, whereas previous association in a single study ranked much lower. High relative probabilities were also observed for location in a functional protein domain, although location in a region evolutionarily conserved in vertebrates was ranked high by the data but not by the experts. Our empirical evidence did not support the importance attributed by the experts to whether the gene encodes a protein in a pathway or shows interactions relevant to the trait. Our findings provide insight into the selection and weighting of different types of knowledge in SNP or gene prioritization, and point to areas requiring further research.

SNP prioritization using a Bayesian probability of association.

Prioritization is the process whereby a set of possible candidate genes or SNPs is ranked so that the most promising can be taken forward into further studies. In a genome-wide association study, prioritization is usually based on the P-values alone, but researchers sometimes take account of external annotation information about the SNPs such as whether the SNP lies close to a good candidate gene. Using external information in this way is inherently subjective and is often not formalized, making the analysis difficult to reproduce. Building on previous work that has identified 14 important types of external information, we present an approximate Bayesian analysis that produces an estimate of the probability of association. The calculation combines four sources of information: the genome-wide data, SNP information derived from bioinformatics databases, empirical SNP weights, and the researchers' subjective prior opinions. The calculation is fast enough that it can be applied to millions of SNPS and although it does rely on subjective judgments, those judgments are made explicit so that the final SNP selection can be reproduced. We show that the resulting probability of association is intuitively more appealing than the P-value because it is easier to interpret and it makes allowance for the power of the study. We illustrate the use of the probability of association for SNP prioritization by applying it to a meta-analysis of kidney function genome-wide association studies and demonstrate that SNP selection performs better using the probability of association compared with P-values alone.

Inertance estimation in a lumped-parameter hydraulic simulator of human circulation.

Pulsatile mock loop systems are largely used to investigate the cardiovascular system in vitro. They consist of a pump, which replicates the heart, coupled with a lumped-parameter hydraulic afterload, which simulates vasculature. An accurate dimensioning of components is required for a reliable mimicking of the physiopathological behavior of the system. However, it is not possible to create a component for the afterload inertance, and inertance contributions are present in the entire circuit. Hence, in the literature, inertance is neglected or qualitatively evaluated. In this paper, we propose two quantitative methods (Maximum-likelihood estimation (MLE) and Bayesian estimation) for estimating afterload inertance based on observed pressure and flow waveforms. These methods are also applied to a real mock loop system. Results show that the system has an inertance comparable with the literature reference value of the entire systemic circulation, and that the expected variations over inlet average flow and pulse frequency are in general confirmed. Comparing the methods, the Bayesian approach results in higher and more stable estimations than the classical MLE.

Multicriteria decision frontiers for prescription anomaly detection over time.

Health care prescription fraud and abuse result in major financial losses and adverse health effects. The growing budget deficits of health insurance programs and recent opioid drug abuse crisis in the United States have accelerated the use of analytical methods. Unsupervised methods such as clustering and anomaly detection could help the health care auditors to evaluate the billing patterns when embedded into rule-based frameworks. These decision models can aid policymakers in detecting potential suspicious activities. This manuscript proposes an unsupervised temporal learning-based decision frontier model using the real world Medicare Part D prescription data collected over 5 years. First, temporal probabilistic hidden groups of drugs are retrieved using a structural topic model with covariates. Next, we construct combined concentration curves and Gini measures considering the weighted impact of temporal observations for prescription patterns, in addition to the Gini values for the cost. The novel decision frontier utilizes this output and enables health care practitioners to assess the trade-offs among different criteria and to identify audit leads.

Underdetection in a stochastic SIR model for the analysis of the COVID-19 Italian epidemic.

We propose a way to model the underdetection of infected and removed individuals in a compartmental model for estimating the COVID-19 epidemic. The proposed approach is demonstrated on a stochastic SIR model, specified as a system of stochastic differential equations, to analyse data from the Italian COVID-19 epidemic. We find that a correct assessment of the amount of underdetection is important to obtain reliable estimates of the critical model parameters. The adaptation of the model in each time interval between relevant government decrees implementing contagion mitigation measures provides short-term predictions and a continuously updated assessment of the basic reproduction number.

Prevalence of Neuropathic Pain in Cancer Patients: Pooled Estimates From a Systematic Review of Published Literature and Results From a Survey Conducted in 50 Italian Palliative Care Centers.

CONTEXT: Because of the increasing body of literature on neuropathic cancer pain (NCP), an accurate estimate of its prevalence requires recurring updates. OBJECTIVES: To provide this estimate using information from a systematic review and a survey. METHODS: Using MEDLINE, Embase, and a previous review, we searched for studies published up to 2014 reporting data on NCP prevalence in adult cancer populations. Pooled prevalence rates from observational prospective studies were computed. The association between NCP prevalence and possible predictors was investigated for oncology and palliative settings. Prevalence rates were extracted from a questionnaire answered by 137 physicians working in 50 Italian centers of palliative care. Estimates from studies conducted in palliative settings and from the experts were analyzed separately and eventually pooled with an informative Bayesian random-effect model. RESULTS: Twenty-nine observational studies were identified. The overall pooled prevalence was 31.2%, with high heterogeneity; similar figures were observed when oncology and palliative settings were individually considered. A slightly higher prevalence of NCP was detected for hospice/inpatients as compared to outpatients, in both settings. The mean NCP prevalence reported by the survey experts was 44.2%; the pooled Bayesian estimate for the palliative setting corresponded to 43.0% (95% CI: 40.0-46.0). The subgroup with the lowest heterogeneity and where the literature and experts' estimates were closest is hospice/inpatients, with a pooled Bayesian prevalence rate of 34.9% (95% CI: 29.9-41.0). CONCLUSION: The systematic review and the survey suggest that more than one in three patients with cancer pain also experiences NCP.

A Rao-Blackwellized particle filter for joint parameter estimation and biomass tracking in a stochastic predator-prey system.

Functional response estimation and population tracking in predator-prey systems are critical problems in ecology. In this paper we consider a stochastic predator-prey system with a Lotka-Volterra functional response and propose a particle filtering method for: (a) estimating the behavioral parameter representing the rate of effective search per predator in the functional response and (b) forecasting the population biomass using field data. In particular, the proposed technique combines a sequential Monte Carlo sampling scheme for tracking the time-varying biomass with the analytical integration of the unknown behavioral parameter. In order to assess the performance of the method, we show results for both synthetic and observed data collected in an acarine predator-prey system, namely the pest mite Tetranychus urticae and the predatory mite Phytoseiulus persimilis.

Nonlinear functional response parameter estimation in a stochastic predator-prey model.

Parameter estimation for the functional response of predator-prey systems is a critical methodological problem in population ecology. In this paper we consider a stochastic predator-prey system with non-linear Ivlev functional response and propose a method for model parameter estimation based on time series of field data. We tackle the problem of parameter estimation using a Bayesian approach relying on a Markov Chain Monte Carlo algorithm. The efficiency of the method is tested on a set of simulated data. Then, the method is applied to a predator-prey system of importance for Integrated Pest Management and biological control, the pest mite Tetranychus urticae and the predatory mite Phytoseiulus persimilis. The model is estimated on a dataset obtained from a field survey. Finally, the estimated model is used to forecast predator-prey dynamics in similar fields, with slightly different initial conditions.

Bayesian inference for functional response in a stochastic predator-prey system.

We present a Bayesian method for functional response parameter estimation starting from time series of field data on predator-prey dynamics. Population dynamics is described by a system of stochastic differential equations in which behavioral stochasticities are represented by noise terms affecting each population as well as their interaction. We focus on the estimation of a behavioral parameter appearing in the functional response of predator to prey abundance when a small number of observations is available. To deal with small sample sizes, latent data are introduced between each pair of field observations and are considered as missing data. The method is applied to both simulated and observational data. The results obtained using different numbers of latent data are compared with those achieved following a frequentist approach. As a case study, we consider an acarine predator-prey system relevant to biological control problems.

Modeling external risks in project management.

To ascertain the viability of a project, undertake resource allocation, take part in bidding processes, and other related decisions, modern project management requires forecasting techniques for cost, duration, and performance of a project, not only under normal circumstances, but also under external events that might abruptly change the status quo. We provide a Bayesian framework that provides a global forecast of a project's performance. We aim at predicting the probabilities and impacts of a set of potential scenarios caused by combinations of disruptive events, and using this information to deal with project management issues. To introduce the methodology, we focus on a project's cost, but the ideas equally apply to project duration or performance forecasting. We illustrate our approach with an example based on a real case study involving estimation of the uncertainty in project cost while bidding for a contract.

A simple approach for improving the hybrid MMVB force field: application to the photoisomerization of s-cis butadiene.

MMVB is a QM/MM hybrid method, consisting of a molecular mechanics force field coupled to a valence bond Heisenberg Hamiltonian parametrized from ab initio CASSCF calculations on several prototype molecules. The Heisenberg Hamiltonian matrix elements Q(ij) and K(ij), whose expressions are partitioned here into a primary contribution and second-order correction terms, are calculated analytically in MMVB. When the original MMVB force field fails to produce potential energy surfaces accurate enough for dynamics calculations, we show that significant improvements can be made by refitting the second-order correction terms for the particular molecule(s) being studied. This "local" reparametrization is based on values of K(ij) extracted (using effective Hamiltonian techniques) from CASSCF calculations on the same molecule(s). The method is demonstrated for the photoisomerization of s-cis butadiene, and we explain how the correction terms that enabled a successful MMVB dynamics study [Garavelli, M.; Bernardi, F.; Olivucci, M.; Bearpark, M. J.; Klein, S.; Robb, M. A. J Phys Chem A 2001, 105, 11496] were refitted.