Modeling Associations Between Latent Event Processes Governing Time Series of Pulsing Hormones

Liu, Huayu; Carlson, Nichole E.; Grunwald, Gary K.; Polotsky, Alex J.

doi:10.1111/biom.12790

Cited by 4 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our decision model is a marked temporal point process (MTPP) (Aalen et al, 2008), which is a natural tool to model discrete events in continuous time. It has been widely applied and become increasingly popular in various domains, including social science (Butts and Marcum, 2017), medical analytics (Liu et al, 2018), finance (Hawkes, 2018), and stochastic optimal control (Tabibian et al, 2019). In our example application of Section 1.1, each follow-up visitation is an event: the visitation time is assumed to be stochastically scheduled according to the probability distribution characterized by the proposed MTPP; and the assigned tacrolimus dosage, when the visitation happens, is treated as the corresponding "mark.…”

Section: Why Use Our Method?mentioning

confidence: 99%

“…Modeling event data with marker information is important to learn the latent mechanisms that govern the observed stochastic event patterns over time in many domains, such as social science (Butts and Marcum, 2017) and medical analytics (Liu et al, 2018). Marked temporal point processes (Aalen et al, 2008) are a general framework for modeling such event data.…”

Section: Modeling Clinical Decisionsmentioning

confidence: 99%

See 1 more Smart Citation

Personalized Dynamic Treatment Regimes in Continuous Time: A Bayesian Approach for Optimizing Clinical Decisions with Timing

et al. 2022

View full text Add to dashboard Cite

Accurate models of clinical actions and their impacts on disease progression are critical for estimating personalized optimal dynamic treatment regimes (DTRs) in medical/health research, especially in managing chronic conditions. Traditional statistical methods for DTRs usually focus on estimating the optimal treatment or dosage at each given medical intervention, but overlook the important question of "when this intervention should happen." We fill this gap by developing a two-step Bayesian approach to optimize clinical decisions with timing. In the first step, we build a generative model for a sequence of medical interventions-which are discrete events in continuous time-with a marked temporal point process (MTPP) where the mark is the assigned treatment or dosage. Then this clinical action model is embedded into a Bayesian joint framework where the other components model clinical observations including longitudinal medical measurements and time-to-event data conditional on treatment histories. In the second step, we propose a policy gradient method to learn the personalized optimal clinical decision that maximizes the patient survival by interacting the MTPP with the model on clinical observations while accounting for uncertainties in clinical observations learned from the posterior inference of the Bayesian joint model in the first step. A signature application of the proposed approach is to schedule follow-up visitations and assign a dosage at each visitation for patients after kidney transplantation. We evaluate our approach with comparison to alternative methods on both simulated and real-world datasets. In our experiments, the personalized decisions made by the proposed method are clinically useful: they are interpretable and successfully help improve patient survival.

show abstract

Section: Why Use Our Method?mentioning

confidence: 99%

Section: Modeling Clinical Decisionsmentioning

confidence: 99%

Personalized Dynamic Treatment Regimes in Continuous Time: A Bayesian Approach for Optimizing Clinical Decisions with Timing

et al. 2022

View full text Add to dashboard Cite

show abstract

“…applied and become increasingly popular in various domains, including social science (Butts and Marcum, 2017), medical analytics (Liu et al, 2018), finance (Hawkes, 2018), and stochastic optimal control (Tabibian et al, 2019). In our example application of section 1.1, each follow-up visitation is an event: the visitation time is assumed to be stochastically scheduled according to the probability distribution characterized by the proposed MTPP; and the assigned tacrolimus dosage, when the visitation happens, is treated as the corresponding "mark.…”

Section: Why Use Our Method?mentioning

confidence: 99%

Section: Modeling Clinical Decisionsmentioning

confidence: 99%

Personalized Dynamic Treatment Regimes in Continuous Time: A Bayesian Approach for Optimizing Clinical Decisions with Timing

Hua

Mei

Zohar

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…The issue of the statistical estimation of signal features becomes even harder when one considers several linked data series, such as joint measurement of LH and FSH levels [50]. To decipher the inherent multi-hormone interactions gonadotropins are part of, Veldhuis et al have combined peak detection algorithms, deconvolution-based methods and biomathematical modeling to reconstruct unobserved signals in a framework called ensemble models [51].…”

Section: 2) Computational Approaches Of the Gnrh And Lh Pulse Generatormentioning

confidence: 99%

Advances in computational modeling approaches of pituitary gonadotropin signaling

Yvinec

Crépieux

Reiter

et al. 2018

Expert Opinion on Drug Discovery

View full text Add to dashboard Cite

Pituitary gonadotropins play an essential and pivotal role in the control of human and animal reproduction within the hypothalamic-pituitary-gonadal (HPG) axis. The computational modeling of pituitary gonadotropin signaling encompasses phenomena of different natures such as the dynamic encoding of gonadotropin secretion, and the intracellular cascades triggered by gonadotropin binding to their cognate receptors, resulting in a variety of biological outcomes. Areas covered: The authors provide an overview of the historical and ongoing issues in modeling and data analysis related to gonadotropin secretion in the field of both physiology and neuroendocrinology. They mention the different mathematical formalisms involved, their interest and limits. They also discuss open statistical questions in signal analysis associated with key endocrine issues and review recent advances in the modeling of the intracellular pathways activated by gonadotropins, which yields promising development for innovative approaches in drug discovery. Expert opinion: The greatest challenge to be tackled in computational modeling of pituitary gonadotropin signaling is the embedding of gonadotropin signaling within its natural multi-scale environment, from the single cell level, to the organic and whole HPG level. The development of modeling approaches of G protein-coupled receptor signaling, together with multicellular systems biology may lead to unexampled mechanistic understanding with critical expected fallouts in the therapeutic management of reproduction.

show abstract

Modeling Associations Between Latent Event Processes Governing Time Series of Pulsing Hormones

Cited by 4 publications

References 23 publications

Personalized Dynamic Treatment Regimes in Continuous Time: A Bayesian Approach for Optimizing Clinical Decisions with Timing

Personalized Dynamic Treatment Regimes in Continuous Time: A Bayesian Approach for Optimizing Clinical Decisions with Timing

Personalized Dynamic Treatment Regimes in Continuous Time: A Bayesian Approach for Optimizing Clinical Decisions with Timing

Advances in computational modeling approaches of pituitary gonadotropin signaling

Contact Info

Product

Resources

About