Spatial cluster detection using dynamic programming

Background

The task of spatial cluster detection involves finding spatial regions where some property deviates from the norm or the expected value. In a probabilistic setting this task can be expressed as finding a region where some event is significantly more likely than usual. Spatial cluster detection is of interest in fields such as biosurveillance, mining of astronomical data, military surveillance, and analysis of fMRI images. In almost all such applications we are interested both in the question of whether a cluster exists in the data, and if it exists, we are interested in finding the most accurate characterization of the cluster.

Methods

We present a general dynamic programming algorithm for grid-based spatial cluster detection. The algorithm can be used for both Bayesian maximum a-posteriori (MAP) estimation of the most likely spatial distribution of clusters and Bayesian model averaging over a large space of spatial cluster distributions to compute the posterior probability of an unusual spatial clustering. The algorithm is explained and evaluated in the context of a biosurveillance application, specifically the detection and identification of Influenza outbreaks based on emergency department visits. A relatively simple underlying model is constructed for the purpose of evaluating the algorithm, and the algorithm is evaluated using the model and semi-synthetic test data.

Results

When compared to baseline methods, tests indicate that the new algorithm can improve MAP estimates under certain conditions: the greedy algorithm we compared our method to was found to be more sensitive to smaller outbreaks, while as the size of the outbreaks increases, in terms of area affected and proportion of individuals affected, our method overtakes the greedy algorithm in spatial precision and recall. The new algorithm performs on-par with baseline methods in the task of Bayesian model averaging.

Conclusions

We conclude that the dynamic programming algorithm performs on-par with other available methods for spatial cluster detection and point to its low computational cost and extendability as advantages in favor of further research and use of the algorithm.

Conditional Anomaly Detection with Soft Harmonic Functions

In this paper, we consider the problem of conditional anomaly detection that aims to identify data instances with an unusual response or a class label. We develop a new non-parametric approach for conditional anomaly detection based on the soft harmonic solution, with which we estimate the confidence of the label to detect anomalous mislabeling. We further regularize the solution to avoid the detection of isolated examples and examples on the boundary of the distribution support. We demonstrate the efficacy of the proposed method on several synthetic and UCI ML datasets in detecting unusual labels when compared to several baseline approaches. We also evaluate the performance of our method on a real-world electronic health record dataset where we seek to identify unusual patient-management decisions.

A Pattern Mining Approach for Classifying Multivariate Temporal Data

We study the problem of learning classification models from complex multivariate temporal data encountered in electronic health record systems. The challenge is to define a good set of features that are able to represent well the temporal aspect of the data. Our method relies on temporal abstractions and temporal pattern mining to extract the classification features. Temporal pattern mining usually returns a large number of temporal patterns, most of which may be irrelevant to the classification task. To address this problem, we present the minimal predictive temporal patterns framework to generate a small set of predictive and non-spurious patterns. We apply our approach to the real-world clinical task of predicting patients who are at risk of developing heparin induced thrombocytopenia. The results demonstrate the benefit of our approach in learning accurate classifiers, which is a key step for developing intelligent clinical monitoring systems.

The application of naive Bayes model averaging to predict Alzheimer's disease from genome-wide data

OBJECTIVE

Predicting patient outcomes from genome-wide measurements holds significant promise for improving clinical care. The large number of measurements (eg, single nucleotide polymorphisms (SNPs)), however, makes this task computationally challenging. This paper evaluates the performance of an algorithm that predicts patient outcomes from genome-wide data by efficiently model averaging over an exponential number of naive Bayes (NB) models.

DESIGN

This model-averaged naive Bayes (MANB) method was applied to predict late onset Alzheimer's disease in 1411 individuals who each had 312,318 SNP measurements available as genome-wide predictive features. Its performance was compared to that of a naive Bayes algorithm without feature selection (NB) and with feature selection (FSNB).

MEASUREMENT

Performance of each algorithm was measured in terms of area under the ROC curve (AUC), calibration, and run time.

RESULTS

The training time of MANB (16.1 s) was fast like NB (15.6 s), while FSNB (1684.2 s) was considerably slower. Each of the three algorithms required less than 0.1 s to predict the outcome of a test case. MANB had an AUC of 0.72, which is significantly better than the AUC of 0.59 by NB (p<0.00001), but not significantly different from the AUC of 0.71 by FSNB. MANB was better calibrated than NB, and FSNB was even better in calibration. A limitation was that only one dataset and two comparison algorithms were included in this study.

CONCLUSION

MANB performed comparatively well in predicting a clinical outcome from a high-dimensional genome-wide dataset. These results provide support for including MANB in the methods used to predict outcomes from large, genome-wide datasets.

A bayesian method for evaluating and discovering disease loci associations

BACKGROUND

A genome-wide association study (GWAS) typically involves examining representative SNPs in individuals from some population. A GWAS data set can concern a million SNPs and may soon concern billions. Researchers investigate the association of each SNP individually with a disease, and it is becoming increasingly commonplace to also analyze multi-SNP associations. Techniques for handling so many hypotheses include the Bonferroni correction and recently developed bayesian methods. These methods can encounter problems. Most importantly, they are not applicable to a complex multi-locus hypothesis which has several competing hypotheses rather than only a null hypothesis. A method that computes the posterior probability of complex hypotheses is a pressing need.

METHODOLOGY/FINDINGS

We introduce the bayesian network posterior probability (BNPP) method which addresses the difficulties. The method represents the relationship between a disease and SNPs using a directed acyclic graph (DAG) model, and computes the likelihood of such models using a bayesian network scoring criterion. The posterior probability of a hypothesis is computed based on the likelihoods of all competing hypotheses. The BNPP can not only be used to evaluate a hypothesis that has previously been discovered or suspected, but also to discover new disease loci associations. The results of experiments using simulated and real data sets are presented. Our results concerning simulated data sets indicate that the BNPP exhibits both better evaluation and discovery performance than does a p-value based method. For the real data sets, previous findings in the literature are confirmed and additional findings are found.

CONCLUSIONS/SIGNIFICANCE

We conclude that the BNPP resolves a pressing problem by providing a way to compute the posterior probability of complex multi-locus hypotheses. A researcher can use the BNPP to determine the expected utility of investigating a hypothesis further. Furthermore, we conclude that the BNPP is a promising method for discovering disease loci associations.

Application of an efficient Bayesian discretization method to biomedical data

Background

Several data mining methods require data that are discrete, and other methods often perform better with discrete data. We introduce an efficient Bayesian discretization (EBD) method for optimal discretization of variables that runs efficiently on high-dimensional biomedical datasets. The EBD method consists of two components, namely, a Bayesian score to evaluate discretizations and a dynamic programming search procedure to efficiently search the space of possible discretizations. We compared the performance of EBD to Fayyad and Irani's (FI) discretization method, which is commonly used for discretization.

Results

On 24 biomedical datasets obtained from high-throughput transcriptomic and proteomic studies, the classification performances of the C4.5 classifier and the naïve Bayes classifier were statistically significantly better when the predictor variables were discretized using EBD over FI. EBD was statistically significantly more stable to the variability of the datasets than FI. However, EBD was less robust, though not statistically significantly so, than FI and produced slightly more complex discretizations than FI.

Conclusions

On a range of biomedical datasets, a Bayesian discretization method (EBD) yielded better classification performance and stability but was less robust than the widely used FI discretization method. The EBD discretization method is easy to implement, permits the incorporation of prior knowledge and belief, and is sufficiently fast for application to high-dimensional data.

Learning Instance-Specific Predictive Models

This paper introduces a Bayesian algorithm for constructing predictive models from data that are optimized to predict a target variable well for a particular instance. This algorithm learns Markov blanket models, carries out Bayesian model averaging over a set of models to predict a target variable of the instance at hand, and employs an instance-specific heuristic to locate a set of suitable models to average over. We call this method the instance-specific Markov blanket (ISMB) algorithm. The ISMB algorithm was evaluated on 21 UCI data sets using five different performance measures and its performance was compared to that of several commonly used predictive algorithms, including nave Bayes, C4.5 decision tree, logistic regression, neural networks, k-Nearest Neighbor, Lazy Bayesian Rules, and AdaBoost. Over all the data sets, the ISMB algorithm performed better on average on all performance measures against all the comparison algorithms.

Conditional outlier detection for clinical alerting

We develop and evaluate a data-driven approach for detecting unusual (anomalous) patient-management actions using past patient cases stored in an electronic health record (EHR) system. Our hypothesis is that patient-management actions that are unusual with respect to past patients may be due to a potential error and that it is worthwhile to raise an alert if such a condition is encountered. We evaluate this hypothesis using data obtained from the electronic health records of 4,486 post-cardiac surgical patients. We base the evaluation on the opinions of a panel of experts. The results support that anomaly-based alerting can have reasonably low false alert rates and that stronger anomalies are correlated with higher alert rates.

An efficient bayesian method for predicting clinical outcomes from genome-wide data

This paper compares the predictive performance and efficiency of several machine-learning methods when applied to a genome-wide dataset on Alzheimer's disease that contains 312,318 SNP measurements on 1411 cases. In particular, a Bayesian algorithm is introduced and compared to several standard machine-learning methods. The results show that the Bayesian algorithm predicts outcomes comparably to the standard methods, and it requires less total training time. These results support the further development and evaluation of the Bayesian algorithm.

A fast algorithm for learning epistatic genomic relationships

Genetic epidemiologists strive to determine the genetic profile of diseases. Epistasis is the interaction between two or more genes to affect phenotype. Due to the often non-linearity of the interaction, it is difficult to detect statistical patterns of epistasis. Combinatorial methods for detecting epistasis investigate a subset of combinations of genes without employing a search strategy. Therefore, they do not scale to handling the high-dimensional data found in genome-wide association studies (GWAS). We represent genome-phenome interactions using a Bayesian network rule, which is a specialized Bayesian network. We develop an efficient search algorithm to learn from data a high scoring rule that may contain two or more interacting genes. Our experimental results using synthetic data indicate that this algorithm detects interacting genes as well as a Bayesian network combinatorial method, and it is much faster. Our results also indicate that the algorithm can successfully learn genome-phenome relationships using a real GWAS dataset.

Identifying Deviations from Usual Medical Care using a Statistical Approach

Developing methods to detect deviations from usual medical care may be useful in the development of automated clinical alerting systems to alert clinicians to treatment choices that warrant additional consideration. We developed a method for identifying deviations in medication administration in the intensive care unit that is based on learning logistic regression models from past patient data that when applied to current patient data identifies statistically unusual treatment decisions. The models predicted a total of 53 deviations for 6 medications on a set of 3000 patient cases. A set of 12 predicted deviations and 12 non-deviations was evaluated by a group of intensive care physicians. Overall, the predicted deviations were assessed to often warrant an alert and to be clinically useful, and furthermore, the frequency with which such alerts would be raised is not likely to be disruptive in a clinical setting.

A Bayesian spatio-temporal method for disease outbreak detection

A system that monitors a region for a disease outbreak is called a disease outbreak surveillance system. A spatial surveillance system searches for patterns of disease outbreak in spatial subregions of the monitored region. A temporal surveillance system looks for emerging patterns of outbreak disease by analyzing how patterns have changed during recent periods of time. If a non-spatial, non-temporal system could be converted to a spatio-temporal one, the performance of the system might be improved in terms of early detection, accuracy, and reliability. A Bayesian network framework is proposed for a class of space-time surveillance systems called BNST. The framework is applied to a non-spatial, non-temporal disease outbreak detection system called PC in order to create the spatio-temporal system called PCTS. Differences in the detection performance of PC and PCTS are examined. The results show that the spatio-temporal Bayesian approach performs well, relative to the non-spatial, non-temporal approach.

The angular selectivity of visual cortical cells to moving gratings

1. Grating patterns were used to obtain a quantitative description of cells in the visual cortex of the cat whose response amplitude depended critically upon the orientation of the moving grating.2. In all such cells the impulse frequency was found to decrease linearly with angle on either side of an optimum angle (the preferred angle) until the response fell to zero or to a base frequency. The angular rate of change of response varied between cells and was expressed as the half-width at half amplitude (the angular selectivity).3. The angular selectivity of thirty-five cells was determined and more than half (nineteen) of these fell within the range 14-26 degrees .4. Fourteen cells responded optimally only when the grating was moved in one direction. Twenty-one cells responded optimally to two directions of movement 180 degrees apart, but the response in the two directions was not always equal.5. No significant correlation was found between the response amplitude at the optimum angle and the angular selectivity.6. The distribution of preferred angles did not show any difference between the oblique orientations and the vertical and horizontal orientations.7. These results are compared with a previous psychophysical estimate of angular selectivity.

Bayesian rule learning for biomedical data mining

MOTIVATION

Disease state prediction from biomarker profiling studies is an important problem because more accurate classification models will potentially lead to the discovery of better, more discriminative markers. Data mining methods are routinely applied to such analyses of biomedical datasets generated from high-throughput 'omic' technologies applied to clinical samples from tissues or bodily fluids. Past work has demonstrated that rule models can be successfully applied to this problem, since they can produce understandable models that facilitate review of discriminative biomarkers by biomedical scientists. While many rule-based methods produce rules that make predictions under uncertainty, they typically do not quantify the uncertainty in the validity of the rule itself. This article describes an approach that uses a Bayesian score to evaluate rule models.

RESULTS

We have combined the expressiveness of rules with the mathematical rigor of Bayesian networks (BNs) to develop and evaluate a Bayesian rule learning (BRL) system. This system utilizes a novel variant of the K2 algorithm for building BNs from the training data to provide probabilistic scores for IF-antecedent-THEN-consequent rules using heuristic best-first search. We then apply rule-based inference to evaluate the learned models during 10-fold cross-validation performed two times. The BRL system is evaluated on 24 published 'omic' datasets, and on average it performs on par or better than other readily available rule learning methods. Moreover, BRL produces models that contain on average 70% fewer variables, which means that the biomarker panels for disease prediction contain fewer markers for further verification and validation by bench scientists.

Bayesian modeling of unknown diseases for biosurveillance

This paper investigates Bayesian modeling of unknown causes of events in the context of disease-outbreak detection. We introduce a Bayesian approach that models and detects both (1) known diseases (e.g., influenza and anthrax) by using informative prior probabilities and (2) unknown diseases (e.g., a new, highly contagious respiratory virus that has never been seen before) by using relatively non-informative prior probabilities. We report the results of simulation experiments which support that this modeling method can improve the detection of new disease outbreaks in a population. A key contribution of this paper is that it introduces a Bayesian approach for jointly modeling both known and unknown causes of events. Such modeling has broad applicability in medical informatics, where the space of known causes of outcomes of interest is seldom complete.

Generalized AMOC curves for evaluation and improvement of event surveillance

We introduce Generalized Activity Monitoring Operating Characteristic (G-AMOC) curves, a new framework for evaluation of outbreak detection systems. G-AMOC curves provide a new approach to evaluating and improving the timeliness of disease outbreak detection by taking the user's response protocol into account and considering when the user will initiate an investigation in response to the system's alerts. The standard AMOC curve is a special case of G-AMOC curves that assumes a trivial response protocol (initiating a new and separate investigation in response to each alert signal). Practical application of a surveillance system is often improved, however, by using more elaborate response protocols, such as grouping alerts or ignoring isolated signals. We present results of experiments demonstrating that we can use G-AMOC curves as 1) a descriptive tool, to provide a more accurate comparison of systems than the standard AMOC curve, and 2) as a prescriptive tool, to choose appropriate response protocols for a detection system, and thus improve its performance.

Analysis of a failed clinical decision support system for management of congestive heart failure

In order to increase compliance with The Joint Commission's Congestive Heart Failure Core Measures, a rule based clinical decision support system (CDSS) was developed and deployed at a community hospital in our health system. We evaluated the performance of the CDSS in identifying patients with primary congestive heart failure (CHF)and identified problems encountered with its introduction. Performance of the CDSS was compared against a manual review of records of patients with diagnosis of primary CHF. The CDSS had a sensitivity of 0.79 and PPV of 0.11. The CDSS issued multiple alerts for majority of the patients(74%). The number of alerts issued for patients without primary CHF was large, and for a majority of patients (63%) physicians did not respond to alerts the first time. The CDSS performed poorly and was eventually withdrawn but provided insight into a subsequently successful method for managing CHF.

Bayesian prediction of an epidemic curve

An epidemic curve is a graph in which the number of new cases of an outbreak disease is plotted against time. Epidemic curves are ordinarily constructed after the disease outbreak is over. However, a good estimate of the epidemic curve early in an outbreak would be invaluable to health care officials. Currently, techniques for predicting the severity of an outbreak are very limited. As far as predicting the number of future cases, ordinarily epidemiologists simply make an educated guess as to how many people might become affected. We develop a model for estimating an epidemic curve early in an outbreak, and we show results of experiments testing its accuracy.

Evaluation of preprocessing techniques for chief complaint classification

OBJECTIVE

To determine whether preprocessing chief complaints before automatically classifying them into syndromic categories improves classification performance.

METHODS

We preprocessed chief complaints using two preprocessors (CCP and EMT-P) and evaluated whether classification performance increased for a probabilistic classifier (CoCo) or for a keyword-based classifier (modification of the NYC Department of Health and Mental Hygiene chief complaint coder (KC)).

RESULTS

CCP exhibited high accuracy (85%) in preprocessing chief complaints but only slightly improved CoCo's classification performance for a few syndromes. EMT-P, which splits chief complaints into multiple problems, substantially increased CoCo's sensitivity for all syndromes. Preprocessing with CCP or EMT-P only improved KC's sensitivity for the Constitutional syndrome.

CONCLUSION

Evaluation of preprocessing systems should not be limited to accuracy of the preprocessor but should include the effect of preprocessing on syndromic classification. Splitting chief complaints into multiple problems before classification is important for CoCo, but other preprocessing steps only slightly improved classification performance for CoCo and a keyword-based classifier.

A statistical approach for detecting deviations from usual medical care

Detecting deviations from usual medical care is valuable in identifying potentially concerning patient management events, both in real time and retrospectively. We describe a statistical method for identification of deviations in medication administration. The preliminary results reported here characterize the statistical properties of the identified deviations. Future research will investigate which deviations are clinically useful.

Evidence-based anomaly detection in clinical domains

Anomaly detection methods can be very useful in identifying interesting or concerning events. In this work, we develop and examine new probabilistic anomaly detection methods that let us evaluate management decisions for a specific patient and identify those decisions that are highly unusual with respect to patients with the same or similar condition. The statistics used in this detection are derived from probabilistic models such as Bayesian networks that are learned from a database of past patient cases. We evaluate our methods on the problem of detection of unusual hospitalization patterns for patients with community acquired pneumonia. The results show very encouraging detection performance with 0.5 precision at 0.53 recall and give us hope that these techniques may provide the basis of intelligent monitoring systems that alert clinicians to the occurrence of unusual events or decisions.

A recursive algorithm for spatial cluster detection

Spatial cluster detection involves finding spatial subregions of some larger region where clusters of some event are occurring. For example, in the case of disease outbreak detection, we want to find clusters of disease cases so as to pinpoint where the outbreak is occurring. When doing spatial cluster detection, we must first articulate the subregions of the region being analyzed. A simple approach is to represent the entire region by an n x n grid. Then we let every subset of cells in the grid represent a subregion. With this representation, the number of subregions is equal to 2(n2) -1. If n is not small, it is intractable to check every subregion. The time complexity of checking all the subregions that are rectangles is (n(4). Neill et al. performed Bayesian spatial cluster detection by only checking every rectangle. In the current paper, we develop a recursive algorithm which searches a richer set of subregions. We provide results of simulation experiments evaluating the detection power and accuracy of the algorithm.

Issues in applied statistics for public health bioterrorism surveillance using multiple data streams: research needs

The objective of this report is to provide a basis to inform decisions about priorities for developing statistical research initiatives in the field of public health surveillance for emerging threats. Rapid information system advances have created a vast opportunity of secondary data sources for information to enhance the situational and health status awareness of populations. While the field of medical informatics and initiatives to standardize healthcare-seeking encounter records continue accelerating, it is necessary to adapt analytic and statistical methodologies to mature in sync with sibling information science technologies. One major right-of-passage for statistical inference is to advance the optimal application of analytic methodologies for using multiple data streams in detecting and characterizing public health population events of importance. This report first describes the problem in general and the data context, then delineates more specifically the practical nature of the problem and the related issues. Approaches currently applied to data with time-series, statistical process control and traditional inference concepts are described with examples in the section on Statistics and the Role of the Analytic Surveillance Data Monitor. These are the techniques that are providing substance to surveillance professionals and enabling use of multiple data streams. The next section describes use of a more complex approach that takes temporal as well as spatial dimensions into consideration for detection and situational awareness regarding event distributions. The space-time statistic has successfully been used to detect and track public health events of interest. Important research questions which are summarized at the end of this report are described in more detail with respect to the methodological application in the respective sections. This was thought to help elucidate the research requirements as summarized later in the report. Following the description of the space-time scan statistical application; this report extends to a less traditional area of promise given what has been observed in recent application of analytic methods. Bayesian networks (BNs) represent a conceptual step with advantages of flexibility for the public health surveillance community. Progression from traditional to the more extending statistical concepts in the context of the dynamic status quo of responsibility and challenge, leads to a conclusion consisting of categorical research needs. The report is structured by design to inform judgment about how to build on practical systems to achieve better analytic outcomes for public health surveillance. There are references to research issues throughout the sections with a summarization at the end, which also includes items previously unmentioned in the report.