Conditional Wasserstein Generator.

The statistical distance of conditional distributions is an essential element of generating target data given some data as in video prediction. We establish how the statistical distances between two joint distributions are related to those between two conditional distributions for three popular statistical distances: f-divergence, Wasserstein distance, and integral probability metrics. Such characterization plays a crucial role in deriving a tractable form of the objective function to learn a conditional generator. For Wasserstein distance, we show that the distance between joint distributions is an upper bound of the expected distance between conditional distributions, and derive a tractable representation of the upper bound. Based on this theoretical result, we propose a new conditional generator, the conditional Wasserstein generator. Our proposed algorithm can be viewed as an extension of Wasserstein autoencoders (Tolstikhin et al. 2018) to conditional generation or as a Wasserstein counterpart of stochastic video generation (SVG) model by Denton and Fergus (Denton et al. 2018). We apply our algorithm to video prediction and video interpolation. Our experiments demonstrate that the proposed algorithm performs well on benchmark video datasets and produces sharper videos than state-of-the-art methods.

Cluster-specific nonignorably missing, endogenous, and continuous regressors in multilevel model for binary outcome.

In multilevel regression models for observational clustered data, regressors can be correlated with cluster-level error components, namely endogenous, due to omitted cluster-level covariates, measurement error, and simultaneity. When endogeneity is ignored, regression coefficient estimators can be severely biased. To deal with endogeneity, instrument variable methods have been widely used. However, the instrument variable method often requires external instrument variables with certain conditions that cannot be verified empirically. Methods that use the within-cluster variations of the endogenous variable work under the restriction that either the outcome or the endogenous variable has a linear relationship with the cluster-level random effect. We propose a new method for binary outcome when it follows a logistic mixed-effects model and the endogenous variable is normally distributed but not linear in the random effect. The proposed estimator capitalizes on the nested data structure without requiring external instrument variables. We show that the proposed estimator is consistent and asymptotically normal. Furthermore, our method can be applied when the endogenous variable is missing in a cluster-specific nonignorable mechanism, without requiring that the missing mechanism be correctly specified. We evaluate the finite sample performance of the proposed approach via simulation and apply the method to a health care study using a San Diego inpatient dataset. Our study demonstrates that the clustered structure can be exploited to draw valid analysis of multilevel data with correlated effects.

ISLES 2016 and 2017-Benchmarking Ischemic Stroke Lesion Outcome Prediction Based on Multispectral MRI.

Performance of models highly depend not only on the used algorithm but also the data set it was applied to. This makes the comparison of newly developed tools to previously published approaches difficult. Either researchers need to implement others' algorithms first, to establish an adequate benchmark on their data, or a direct comparison of new and old techniques is infeasible. The Ischemic Stroke Lesion Segmentation (ISLES) challenge, which has ran now consecutively for 3 years, aims to address this problem of comparability. ISLES 2016 and 2017 focused on lesion outcome prediction after ischemic stroke: By providing a uniformly pre-processed data set, researchers from all over the world could apply their algorithm directly. A total of nine teams participated in ISLES 2015, and 15 teams participated in ISLES 2016. Their performance was evaluated in a fair and transparent way to identify the state-of-the-art among all submissions. Top ranked teams almost always employed deep learning tools, which were predominately convolutional neural networks (CNNs). Despite the great efforts, lesion outcome prediction persists challenging. The annotated data set remains publicly available and new approaches can be compared directly via the online evaluation system, serving as a continuing benchmark (www.isles-challenge.org).

Cox model with interval-censored covariate in cohort studies.

In cohort studies the outcome is often time to a particular event, and subjects are followed at regular intervals. Periodic visits may also monitor a secondary irreversible event influencing the event of primary interest, and a significant proportion of subjects develop the secondary event over the period of follow-up. The status of the secondary event serves as a time-varying covariate, but is recorded only at the times of the scheduled visits, generating incomplete time-varying covariates. While information on a typical time-varying covariate is missing for entire follow-up period except the visiting times, the status of the secondary event are unavailable only between visits where the status has changed, thus interval-censored. One may view interval-censored covariate of the secondary event status as missing time-varying covariates, yet missingness is partial since partial information is provided throughout the follow-up period. Current practice of using the latest observed status produces biased estimators, and the existing missing covariate techniques cannot accommodate the special feature of missingness due to interval censoring. To handle interval-censored covariates in the Cox proportional hazards model, we propose an available-data estimator, a doubly robust-type estimator as well as the maximum likelihood estimator via EM algorithm and present their asymptotic properties. We also present practical approaches that are valid. We demonstrate the proposed methods using our motivating example from the Northern Manhattan Study.

Evaluation of a technology-enhanced integrated care model for frail older persons: protocol of the SPEC study, a stepped-wedge cluster randomized trial in nursing homes.

BACKGROUND: Limited evidence exists on the effectiveness of the chronic care model for people with multimorbidity. This study aims to evaluate the effectiveness of an information and communication technology- (ICT-)enhanced integrated care model, called Systems for Person-centered Elder Care (SPEC), for frail older adults at nursing homes. METHODS/DESIGN: SPEC is a prospective stepped-wedge cluster randomized trial conducted at 10 nursing homes in South Korea. Residents aged 65 or older meeting the inclusion/exclusion criteria in all the homes are eligible to participate. The multifaceted SPEC intervention, a geriatric care model guided by the chronic care model, consists of five components: comprehensive geriatric assessment for need/risk profiling, individual need-based care planning, interdisciplinary case conferences, person-centered care coordination, and a cloud-based information and communications technology (ICT) tool supporting the intervention process. The primary outcome is quality of care for older residents using a composite measure of quality indicators from the interRAI LTCF assessment system. Outcome assessors and data analysts will be blinded to group assignment. Secondary outcomes include quality of life, healthcare utilization, and cost. Process evaluation will be also conducted. DISCUSSION: This study is expected to provide important new evidence on the effectiveness, cost-effectiveness, and implementation process of an ICT-supported chronic care model for older persons with multiple chronic illnesses. The SPEC intervention is also unique as the first registered trial implementing an integrated care model using technology to promote person-centered care for frail older nursing home residents in South Korea, where formal LTC was recently introduced. TRIAL REGISTRATION: ISRCTN11972147.

Cox proportional hazards models with left truncation and time-varying coefficient: Application of age at event as outcome in cohort studies.

When analyzing time-to-event cohort data, two different ways of choosing a time scale have been discussed in the literature: time-on-study or age at onset of disease. One advantage of choosing the latter is interpretability of the hazard ratio as a function of age. To handle the analysis of age at onset in a principled manner, we present an analysis of the Cox Proportional Hazards model with time-varying coefficient for left-truncated and right-censored data. In the analysis of Northern Manhattan Study (NOMAS) with age at onset of stroke as outcome, we demonstrate that well-established risk factors may be important only around a certain age span and less established risk factors can have a strong effect in a certain age span.

Calibrated propensity score method for survey nonresponse in cluster sampling.

Weighting adjustment is commonly used in survey sampling to correct for unit nonresponse. In cluster sampling, the missingness indicators are often correlated within clusters and the response mechanism is subject to cluster-specific nonignorable missingness. Based on a parametric working model for the response mechanism that incorporates cluster-specific nonignorable missingness, we propose a method of weighting adjustment. We provide a consistent estimator of the mean or totals in cases where the study variable follows a generalized linear mixed-effects model. The proposed method is robust in the sense that the consistency of the estimator does not require correct specification of the functional forms of the response and outcome models. A consistent variance estimator based on Taylor linearization is also proposed. Numerical results, including a simulation and a real-data application, are presented.

Predictors and outcomes of unplanned readmission to a different hospital.

OBJECTIVES: To examine patient, hospital and market factors and outcomes associated with readmission to a different hospital compared with the same hospital. DESIGN: A population-based, secondary analysis using multilevel causal modeling. SETTING: Acute care hospitals in California in the USA. PARTICIPANTS: In total, 509 775 patients aged 50 or older who were discharged alive from acute care hospitals (index hospitalizations), and 59 566 who had a rehospitalization within 30 days following their index discharge. INTERVENTION: No intervention. MAIN OUTCOME MEASURE(S): Thirty-day unplanned readmissions to a different hospital compared with the same hospital and also the costs and health outcomes of the readmissions. RESULTS: Twenty-one percent of patients with a rehospitalization had a different-hospital readmission. Compared with the same-hospital readmission group, the different-hospital readmission group was more likely to be younger, male and have a lower income. The index hospitals of the different-hospital readmission group were more likely to be smaller, for-profit hospitals, which were also more likely to be located in counties with higher competition. The different-hospital readmission group had higher odds for in-hospital death (8.1 vs. 6.7%; P < 0.0001) and greater readmission hospital costs ($15 671.8 vs. $14 286.4; P < 0.001) than the same-hospital readmission group. CONCLUSIONS: Patient, hospital and market characteristics predicted different-hospital readmissions compared with same-hospital readmissions. Mortality and cost outcomes were worse among patients with different-hospital readmissions. Strategies for better care coordination targeting people at risk for different-hospital readmissions are necessary.

Infectious burden and cognitive function: the Northern Manhattan Study.

OBJECTIVE: We hypothesized that infectious burden (IB), a composite serologic measure of exposure to common pathogens (i.e., Chlamydia pneumoniae, Helicobacter pylori, cytomegalovirus, and herpes simplex virus 1 and 2) associated with vascular risk in the prospective Northern Manhattan Study (NOMAS), would also be associated with cognition. METHODS: Cognition was assessed using the Mini-Mental State Examination (MMSE) at enrollment and the modified Telephone Interview for Cognitive Status (TICS-m) at annual follow-up visits. Adjusted linear and logistic regressions were used to measure the association between IB index and MMSE. Generalized estimating equation models were used to evaluate associations with TICS-m and its change over time. RESULTS: Serologies and cognitive assessments were available in 1,625 participants of the NOMAS cohort. In unadjusted analyses, higher IB index was associated with worse cognition (change per standard deviation [SD] of IB for MMSE was -0.77, p < 0.0001, and for first measurements of TICS-m was -1.89, p < 0.0001). These effects were attenuated after adjusting for risk factors (for MMSE adjusted change per SD of IB = -0.17, p = 0.06, for TICS-m adjusted change per SD IB = -0.68, p < 0.0001). IB was associated with MMSE ≤24 (compared to MMSE >24, adjusted odds ratio 1.26 per SD of IB, 95% confidence interval 1.06-1.51). IB was not associated with cognitive decline over time. The results were similar when IB was limited to viral serologies only. CONCLUSION: A measure of IB associated with stroke risk and atherosclerosis was independently associated with cognitive performance in this multiethnic cohort. Past infections may contribute to cognitive impairment.

A marginal approach to reduced-rank penalized spline smoothing with application to multilevel functional data.

Multilevel functional data is collected in many biomedical studies. For example, in a study of the effect of Nimodipine on patients with subarachnoid hemorrhage (SAH), patients underwent multiple 4-hour treatment cycles. Within each treatment cycle, subjects' vital signs were reported every 10 minutes. This data has a natural multilevel structure with treatment cycles nested within subjects and measurements nested within cycles. Most literature on nonparametric analysis of such multilevel functional data focus on conditional approaches using functional mixed effects models. However, parameters obtained from the conditional models do not have direct interpretations as population average effects. When population effects are of interest, we may employ marginal regression models. In this work, we propose marginal approaches to fit multilevel functional data through penalized spline generalized estimating equation (penalized spline GEE). The procedure is effective for modeling multilevel correlated generalized outcomes as well as continuous outcomes without suffering from numerical difficulties. We provide a variance estimator robust to misspecification of correlation structure. We investigate the large sample properties of the penalized spline GEE estimator with multilevel continuous data and show that the asymptotics falls into two categories. In the small knots scenario, the estimated mean function is asymptotically efficient when the true correlation function is used and the asymptotic bias does not depend on the working correlation matrix. In the large knots scenario, both the asymptotic bias and variance depend on the working correlation. We propose a new method to select the smoothing parameter for penalized spline GEE based on an estimate of the asymptotic mean squared error (MSE). We conduct extensive simulation studies to examine property of the proposed estimator under different correlation structures and sensitivity of the variance estimation to the choice of smoothing parameter. Finally, we apply the methods to the SAH study to evaluate a recent debate on discontinuing the use of Nimodipine in the clinical community.

Reweighting estimators for Cox regression with missing covariate data: analysis of insulin resistance and risk of stroke in the Northern Manhattan Study.

Incomplete covariates often obscure analysis results from a Cox regression. In an analysis of the Northern Manhattan Study (NOMAS) to determine the influence of insulin resistance on the incidence of stroke in nondiabetic individuals, insulin level is unknown for 34.1% of the subjects. The available data suggest that the missingness mechanism depends on outcome variables, which may generate biases in estimating the parameters of interest if only using the complete observations. This article aimed to introduce practical strategies to analyze the NOMAS data and present sensitivity analyses by using the reweighting method in standard statistical packages. When the data set structure is in counting process style, the reweighting estimates can be obtained by built-in procedures with variance estimated by the jackknife method. Simulation results indicate that the jackknife variance estimate provides reasonable coverage probability in moderate sample sizes. We subsequently conducted sensitivity analyses for the NOMAS data, showing that the risk estimates are robust to a variety of missingness mechanisms. At the end of this article, we present the core SAS and R programs used in the analysis.

Heritability Estimation using Regression Models for Correlation.

Heritability estimates a polygenic effect on a trait for a population. Reliable interpretation of heritability is critical in planning further genetic studies to locate a gene responsible for the trait. This study accommodates both single and multiple trait cases by employing regression models for correlation parameter to infer the heritability. Sharing the properties of regression approach, the proposed methods are exible to incorporate non-genetic and/or non-additive genetic information in the analysis. The performances of the proposed model are compared with those using the likelihood approach through simulations and carotid Intima Media Thickness analysis from Northern Manhattan family Study.

HANDLING MISSING DATA BY DELETING COMPLETELY OBSERVED RECORDS.

When data are missing, analyzing records that are completely observed may cause bias or inefficiency. Existing approaches in handling missing data include likelihood, imputation and inverse probability weighting. In this paper, we propose three estimators inspired by deleting some completely observed data in the regression setting. First, we generate artificial observation indicators that are independent of outcome given the observed data and draw inferences conditioning on the artificial observation indicators. Second, we propose a closely related weighting method. The proposed weighting method has more stable weights than those of the inverse probability weighting method (Zhao and Lipsitz, 1992). Third, we improve the efficiency of the proposed weighting estimator by subtracting the projection of the estimating function onto the nuisance tangent space. When data are missing completely at random, we show that the proposed estimators have asymptotic variances smaller than or equal to the variance of the estimator obtained from using completely observed records only. Asymptotic relative efficiency computation and simulation studies indicate that the proposed weighting estimators are more efficient than the inverse probability weighting estimators under wide range of practical situations especially when when the missingness proportion is large.

Handling missing responses in generalized linear mixed model without specifying missing mechanism.

In longitudinal studies, missingness of data is often unavoidable. Valid estimators from the generalized linear mixed model usually rely on the correct specification of the missing data mechanism. An incorrectly specified missing mechanism may lead to a biased estimator. In this article, we propose a class of unbiased estimating equations using pairwise conditional technique to deal with the generalized linear mixed model under benign non-ignorable missingness where specification of the missing model is not needed. We show that the proposed estimator is consistent and asymptotically normal under certain conditions. Simulation results and an example using longitudinal course of neuropsychological data are also shown.

Genotype-adjusted familial correlation analysis using three generalized estimating equations.

We analyze familial correlation of a memory score from Caribbean Hispanic families that have multiple family members affected with Alzheimer's disease, adjusting for having at least one APOE-epsilon4 allele, as well as other confounders. To enhance the efficiency of correlation model, this paper proposes an alternative approach for three generalized estimating equations proposed by Yan and Fine (Statist. Med. 2004; 23:859-874). The efficiency of correlation model is evaluated through the asymptotic relative efficiency computation and simulations.

Carotid plaque surface irregularity predicts ischemic stroke: the northern Manhattan study.

BACKGROUND AND PURPOSE: There is scant population-based evidence regarding extracranial carotid plaque surface irregularity and ischemic stroke. Using a prospective cohort design, we evaluated the association of carotid plaque surface irregularity and the risk of ischemic stroke in a multiethnic population. METHODS: High-resolution B-mode ultrasound of the carotid arteries was performed in 1939 stroke-free subjects (mean age 69+/-10.0 years; 59% women; 53% Hispanic, 25% black, 22% white). Plaque was defined as a focal protrusion 50% greater than the surrounding area and localized along the extracranial carotid tree (internal carotid artery/bifurcation vs common carotid artery). Plaque surface was categorized as regular or irregular. Cox proportional hazard models were used to assess the association of surface characteristics and the risk of ischemic stroke. RESULTS: Among 1939 total subjects, carotid plaque was visualized in 56.3% (1 plaque: 21.6%, >1 plaque: 34.7%, irregular plaque: 5.5%). During a mean follow up of 6.2 years after ultrasound examination, 69 ischemic strokes occurred. Unadjusted cumulative 5-year risks of ischemic stroke were: 1.3%, 3.0%, and 8.5% for no plaque, regular plaque, and irregular plaque, respectively. After adjusting for demographics, traditional vascular risk factors, degree of stenosis, and plaque thickness, presence of irregular plaque (vs no plaque) was independently associated with ischemic stroke (Hazard ratio, 3.1; 95% CI, 1.1 to 8.5). CONCLUSIONS: The presence of irregular carotid plaque independently predicted ischemic stroke in a multiethnic cohort. Plaque surface irregularities assessed by B-mode ultrasonography may help identify intermediate- to high-risk individuals beyond their vascular risk assessed by the presence of traditional risk factors.