Artificial intelligence framework for simulating clinical decision-making: A Markov decision process approach

Bennett, Casey C.; Hauser, Kris

doi:10.1016/j.artmed.2012.12.003

Cited by 286 publications

(139 citation statements)

References 32 publications

Supporting

Mentioning

135

Contrasting

Unclassified

Order By: Relevance

“…Gaweda et al [74, 75] discussed the use of control theoretic approaches to anemia management in patients with end-stage renal disease. Bennett and Hauser [76] discussed a framework for simulating clinical decision making from electronic medical records data. In summary, while the dynamical systems approaches to develop DTRs are emerging, from a statistical perspective they still lag behind the other approaches presented earlier; hence this area is ripe for further development.…”

Section: Discussionmentioning

confidence: 99%

Dynamic Treatment Regimes

Chakraborty

Murphy

2014

Annu. Rev. Stat. Appl.

194

152

View full text Add to dashboard Cite

A dynamic treatment regime consists of a sequence of decision rules, one per stage of intervention, that dictate how to individualize treatments to patients based on evolving treatment and covariate history. These regimes are particularly useful for managing chronic disorders, and fit well into the larger paradigm of personalized medicine. They provide one way to operationalize a clinical decision support system. Statistics plays a key role in the construction of evidence-based dynamic treatment regimes – informing best study design as well as efficient estimation and valid inference. Due to the many novel methodological challenges it offers, this area has been growing in popularity among statisticians in recent years. In this article, we review the key developments in this exciting field of research. In particular, we discuss the sequential multiple assignment randomized trial designs, estimation techniques like Q-learning and marginal structural models, and several inference techniques designed to address the associated non-standard asymptotics. We reference software, whenever available. We also outline some important future directions.

show abstract

Section: Discussionmentioning

confidence: 99%

Dynamic Treatment Regimes

Chakraborty

Murphy

2014

Annu. Rev. Stat. Appl.

194

152

View full text Add to dashboard Cite

show abstract

“…Whilst important to harvest new prospects (such as, for example, the role of artificial intelligence to diagnose disease, which has been shown to outperform humans in patient outcomes and cost),26 it is also important to concentrate on optimising existing technologies. This is particularly true in light of the findings that socio-economic returns from complex health technologies are likely to take a long time to materialise 27.…”

Section: Discussionmentioning

confidence: 99%

Qualitative analysis of multi-disciplinary round-table discussions on the acceleration of benefits and data analytics through hospital electronic prescribing (ePrescribing) systems

Cresswell

Coleman

Smith

et al. 2016

jhi

View full text Add to dashboard Cite

A more coherent national approach to realising financial benefits from ePrescribing systems is needed as implementations progress and the range of tools to collect information will lead to exponential data growth. The move towards more sophisticated closed-loop systems that integrate prescribing, administration and dispensing, as well as increasingly empowered patients accessing their data through portals and portable devices, will accelerate these developments. Meaningful analysis of data will be the key to realise benefits associated with systems.

show abstract

“…In [9], the authors proposed an MDP-based solution to obtain a profitable balance between maintenance cost and network performance, by replacing the failed nodes efficiently. Bennett and Hauser [10] developed a general purpose computational/artificial intelligence framework to address a simulation environment for exploring various healthcare policies, and payment methodologies.…”

Section: Related Workmentioning

confidence: 99%

Prioritized payload tuning mechanism for wireless body area network-based healthcare systems

Moulik

Misra

Chakraborty

et al. 2014

2014 IEEE Global Communications Conference

View full text Add to dashboard Cite

This paper presents a priority-based MAC-frame payload tuning mechanism with reduced energy consumption for healthcare systems that use Wireless Body Area Networks (WBANs). A fundamental problem in WBANs is to prioritize the physiological sensors depending on several health and external criteria. The challenge is to design a dynamic decision making model that can optimize the energy consumption of each physiological sensor. To address this problem we employ the concept of Fuzzy Inference System (FIS) in order to calculate Criticality Index (CI), which signifies the severity or the priority of the physiological data sensed by each sensor. Considering the obtained CI value we proceed with designing a Markov Decision Process (MDP) based dynamic decision making model in order to tune MAC-frame payload by optimizing the energy consumption of each sensor node. We achieve around 25% decrease in the overall energy consumption using our proposed mechanism.

show abstract

Artificial intelligence framework for simulating clinical decision-making: A Markov decision process approach

Cited by 286 publications

References 32 publications

Dynamic Treatment Regimes

Dynamic Treatment Regimes

Qualitative analysis of multi-disciplinary round-table discussions on the acceleration of benefits and data analytics through hospital electronic prescribing (ePrescribing) systems

Prioritized payload tuning mechanism for wireless body area network-based healthcare systems

Contact Info

Product

Resources

About