Joint hierarchical Gaussian process model with application to personalized prediction in medical monitoring

Duan, Leo L.; Wang, Xia; Clancy, John; Szczesniak, Rhonda

doi:10.1002/sta4.178

Cited by 7 publications

(8 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the CF context, the parameter

bold-italicc

indicates the association between the continuous measurements and the risk probability of binary longitudinal measurements. As applied in previous studies (Andrinopoulou and Rizopoulos, 2016; Duan et al., 2018), the prior on c facilitates parameter estimation.…”

Section: Joint Hierarchical Gaussian Process Model With Flexible Link Functionsmentioning

confidence: 99%

See 1 more Smart Citation

Flexible link functions in a joint hierarchical Gaussian process model

Wang

Szczesniak

2020

Biometrics

Self Cite

View full text Add to dashboard Cite

Many longitudinal studies often require jointly modeling a biomarker and an event outcome, in order to provide more accurate inference and dynamic prediction of disease progression. Cystic fibrosis (CF) studies have illustrated the benefits of these models, primarily examining the joint evolution of lung‐function decline and survival. We propose a novel joint model within the shared‐parameter framework that accommodates nonlinear lung‐function trajectories, in order to provide more accurate inference on lung‐function decline over time and to examine the association between evolution of lung function and risk of a pulmonary exacerbation (PE) event recurrence. Specifically, a two‐level Gaussian process (GP) is used to estimate the nonlinear longitudinal trajectories and a flexible link function is introduced for a more accurate depiction of the binary process on the event outcome. Bayesian model assessment is used to evaluate each component of the joint model in simulation studies and an application to longitudinal data on patients receiving care from a CF center. A nonlinear structure is suggested by both longitudinal continuous and binary evaluations. Including a flexible link function improves model fit to these data. The proposed hierarchical GP model with a flexible power link function where Laplace distribution is the baseline (spep) has the best fit of all joint models considered, characterizing how accelerated lung‐function decline corresponds to increased odds of experiencing another PE.

show abstract

“…In the CF context, the parameter

bold-italicc

Section: Joint Hierarchical Gaussian Process Model With Flexible Link Functionsmentioning

confidence: 99%

“…Joint modeling (Rizopoulos, 2012; Kim and Albert, 2016; Duan et al., 2018) is widely used nowadays in a variety of longitudinal studies. Different types of longitudinal measurements are usually collected simultaneously in the studies.…”

Section: Introductionmentioning

confidence: 99%

Flexible link functions in a joint hierarchical Gaussian process model

Wang

Szczesniak

2020

Biometrics

Self Cite

View full text Add to dashboard Cite

show abstract

“…Accounted for complications experienced by patients during follow-up. Multivariate modelling using Gaussian processes [117]. Multivariate modelling and informative processes [76] Employed to classify repeated measurements of hormone levels in early pregnancy to predict pregnancy success in the context of in vitro fertilization [54].…”

Section: Generalised Estimating Equationsmentioning

confidence: 99%

Harnessing repeated measurements of predictor variables for clinical risk prediction: a review of existing methods

et al. 2020

View full text Add to dashboard Cite

Background: Clinical prediction models (CPMs) predict the risk of health outcomes for individual patients. The majority of existing CPMs only harness cross-sectional patient information. Incorporating repeated measurements, such as those stored in electronic health records, into CPMs may provide an opportunity to enhance their performance. However, the number and complexity of methodological approaches available could make it difficult for researchers to explore this opportunity. Our objective was to review the literature and summarise existing approaches for harnessing repeated measurements of predictor variables in CPMs, primarily to make this field more accessible for applied researchers. Methods: MEDLINE, Embase and Web of Science were searched for articles reporting the development of a multivariable CPM for individual-level prediction of future binary or time-to-event outcomes and modelling repeated measurements of at least one predictor. Information was extracted on the following: the methodology used, its specific aim, reported advantages and limitations, and software available to apply the method. Results: The search revealed 217 relevant articles. Seven methodological frameworks were identified: time-dependent covariate modelling, generalised estimating equations, landmark analysis, two-stage modelling, joint-modelling, trajectory classification and machine learning. Each of these frameworks satisfies at least one of three aims: to better represent the predictor-outcome relationship over time, to infer a covariate value at a pre-specified time and to account for the effect of covariate change. Conclusions: The applicability of identified methods depends on the motivation for including longitudinal information and the method's compatibility with the clinical context and available patient data, for both model development and risk estimation in practice.

show abstract

“…This two-level joint model structure, also called JHGP model, is introduced in Duan et al. 10 It includes both population and individual level GPs: at the population level, the bold-italicμy and bold-italicμR monitor trends over time t across all patients; at the individual level, the ψ it capture subject-specific trajectory and is also the shared component between two submodels with the corresponding coefficient parameter c . This assumes that the longitudinal continuous and binary processes are associated at each time t for the i th subject.…”

Section: Joint Model With Gaussian Process and Flexible Link Functionsmentioning

confidence: 99%

“…The aforementioned applications of joint models for simultaneously modeling longitudinal FEV 1 and PEs in CF typically feature FEV 1 as a random-intercepts submodel and PE occurrence through a survival submodel. Although individual improvements have been made in these models, such as including hierarchical GPs in each submodel, 10 there are two aspects that remain largely unexplored. In addition to inclusion of hierarchical GPs in both submodels, we propose an alternative framework motivated by the CF context and the need for additional theoretical investigation.…”

Section: Introductionmentioning

confidence: 99%

Risk factor identification in cystic fibrosis by flexible hierarchical joint models

Wang

Szczesniak

2020

Stat Methods Med Res

Self Cite

View full text Add to dashboard Cite

Cystic fibrosis (CF) is a lethal autosomal disease hallmarked by respiratory failure. Maintaining lung function and minimizing frequency of acute respiratory events known as pulmonary exacerbations are essential to survival. Jointly modeling longitudinal lung function and exacerbation occurrences may provide better inference. We propose a shared-parameter joint hierarchical Gaussian process model with flexible link function to investigate the impacts of both demographic and time-varying clinical risk factors on lung function decline and to examine the associations between lung function and occurrence of pulmonary exacerbation. A two-level Gaussian process is used to capture the nonlinear longitudinal trajectory, and a flexible link function is introduced to the joint model in order to analyze binary process. Bayesian model assessment criteria are provided in examining the overall performance in joint models and marginal fitting in each submodel. We conduct simulation studies and apply the proposed model in a local CF center cohort. In the CF application, a nonlinear structure is supported in modeling both the longitudinal continuous and binary processes. A negative association is detected between lung function and pulmonary exacerbation by the joint model. The importance of risk factors, including gender, diagnostic status, insurance status, and BMI, is examined in joint models.

show abstract

Joint hierarchical Gaussian process model with application to personalized prediction in medical monitoring

Cited by 7 publications

References 30 publications

Flexible link functions in a joint hierarchical Gaussian process model

Flexible link functions in a joint hierarchical Gaussian process model

Harnessing repeated measurements of predictor variables for clinical risk prediction: a review of existing methods

Risk factor identification in cystic fibrosis by flexible hierarchical joint models

Contact Info

Product

Resources

About