Decision analysis networks

Díez, F. J.; Luque, Manuel; Bermejo, Iñigo

doi:10.1016/j.ijar.2018.02.007

Cited by 14 publications

(5 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, OpenMarkov has novel types of PGMs, such as Markov influence diagrams [14] and decision analysis networks [15], developed by our research group, as well as new algorithms for cost-effectiveness analysis with these models [14,39,40]. They can be very useful for teaching health technology assessment (HTA), but that topic is out of the scope of this paper.…”

Section: Discussionmentioning

confidence: 99%

“…OpenMarkov offers support for editing and evaluating several types of PGMs, such as BNs, IDs, Markov IDs [14], and decision analysis networks [15]. It can also edit limitedmemory IDs (LIMIDs) [16] and several types of temporal models, such as dynamic Bayesian networks [17], which include Markov chains and hidden Markov models as a particular case, factored Markov decision processes (MDPs) [18], factored partially observable MDPS (POMDPs) [19], and DLIMIDs [20].…”

Section: Openmarkovmentioning

confidence: 99%

See 1 more Smart Citation

Teaching Probabilistic Graphical Models with OpenMarkov

et al. 2022

Self Cite

View full text Add to dashboard Cite

OpenMarkov is an open-source software tool for probabilistic graphical models. It has been developed especially for medicine, but has also been used to build applications in other fields and for tuition, in more than 30 countries. In this paper we explain how to use it as a pedagogical tool to teach the main concepts of Bayesian networks and influence diagrams, such as conditional dependence and independence, d-separation, Markov blankets, explaining away, optimal policies, expected utilities, etc., and some inference algorithms: logic sampling, likelihood weighting, and arc reversal. The facilities for learning Bayesian networks interactively can be used to illustrate step by step the performance of the two basic algorithms: search-and-score and PC.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Openmarkovmentioning

confidence: 99%

Teaching Probabilistic Graphical Models with OpenMarkov

et al. 2022

Self Cite

View full text Add to dashboard Cite

show abstract

“…We may allow partial orderings of the decisions by building a decision analysis network (DAN) [ 50 ] instead of an ID. However, the main difficulty would be collecting clinical data to elicit these parameters.…”

Section: Discussionmentioning

confidence: 99%

Optimal sequence of tests for the mediastinal staging of non-small cell lung cancer

Luque

Díez

Disdier

2016

BMC Med Inform Decis Mak

Self Cite

View full text Add to dashboard Cite

BackgroundNon-small cell lung cancer (NSCLC) is the most prevalent type of lung cancer and the most difficult to predict. When there are no distant metastases, the optimal therapy depends mainly on whether there are malignant lymph nodes in the mediastinum. Given the vigorous debate among specialists about which tests should be used, our goal was to determine the optimal sequence of tests for each patient.MethodsWe have built an influence diagram (ID) that represents the possible tests, their costs, and their outcomes. This model is equivalent to a decision tree containing millions of branches. In the first evaluation, we only took into account the clinical outcomes (effectiveness). In the second, we used a willingness-to-pay of € 30,000 per quality adjusted life year (QALY) to convert economic costs into effectiveness. We assigned a second-order probability distribution to each parameter in order to conduct several types of sensitivity analysis.ResultsTwo strategies were obtained using two different criteria. When considering only effectiveness, a positive computed tomography (CT) scan must be followed by a transbronchial needle aspiration (TBNA), an endobronchial ultrasound (EBUS), and an endoscopic ultrasound (EUS). When the CT scan is negative, a positron emission tomography (PET), EBUS, and EUS are performed. If the TBNA or the PET is positive, then a mediastinoscopy is performed only if the EBUS and EUS are negative. If the TBNA or the PET is negative, then a mediastinoscopy is performed only if the EBUS and the EUS give contradictory results. When taking into account economic costs, a positive CT scan is followed by a TBNA; an EBUS is done only when the CT scan or the TBNA is negative.This recommendation of performing a TBNA in certain cases should be discussed by the pneumology community because TBNA is a cheap technique that could avoid an EBUS, an expensive test, for many patients.ConclusionsWe have determined the optimal sequence of tests for the mediastinal staging of NSCLC by considering sensitivity, specificity, and the economic cost of each test. The main novelty of our study is the recommendation of performing TBNA whenever the CT scan is positive. Our model is publicly available so that different experts can populate it with their own parameters and re-examine its conclusions. It is therefore proposed as an evidence-based instrument for reaching a consensus.

show abstract

“…Influence diagrams are representations where there is at least one node representing a decision that can be chosen by the decision makers. Another node represents the utility associated with certain outcomes [69][70][71]. In the present case, an influence diagram was developed from the primary data derived from AIS data.…”

Section: Influence Diagramsmentioning

confidence: 99%

A Bayesian Approach to Infer the Sustainable Use of Artificial Reefs in Fisheries and Recreation

Ramos,

Drakeford,

Madiedo

et al. 2024

Sustainability

View full text Add to dashboard Cite

The presence of artificial reefs (ARs) in the south of Portugal that were deployed a few decades ago and the corroboration of fishing patterns and other activities related to the use of these habitats have not been followed. It is important to note that monitoring the use of ARs was difficult in the past but is currently facilitated by the application of non-intrusive tools. In the present study, an approach is developed where, based on monitoring data from fishing and non-fishing boats, influence diagrams (IDs) are constructed to provide some evidence on fisheries or other use patterns and consequent AR effectiveness as coastal tools. These IDs allow us to infer various usefulness scenarios, namely catches, which are tangible, and satisfaction, which is intangible, and overall assessment of ARs and nearby areas in terms of human activities. After calibrating the Bayesian ID based on monitoring evidence, the obtained model was evaluated for several scenarios. In the base case, which assumes the occurrence of more fishing than recreation (assuming 3:1, respectively), the obtained utility is 18.64% (catches) and 31.96% (satisfaction). Of the scenarios run, the one that obtained the best results in the utility nodes together was the second one. The use of these tailored tools and approaches seems to be of fundamental importance for the adequate management of coastal infrastructures, particularly with regard to the inference of fishing resources and their sustainable use. An adequate interpretation based on the use of these tools implies being able to safeguard the ecological balance and economic sustainability of the communities operating in these areas.

show abstract

Decision analysis networks

Cited by 14 publications

References 28 publications

Teaching Probabilistic Graphical Models with OpenMarkov

Teaching Probabilistic Graphical Models with OpenMarkov

Optimal sequence of tests for the mediastinal staging of non-small cell lung cancer

A Bayesian Approach to Infer the Sustainable Use of Artificial Reefs in Fisheries and Recreation

Contact Info

Product

Resources

About