Model-based Decision Research Articles

Disruption of model-based decision making by silencing of serotonin neurons in the dorsal raphe nucleus

Adapting to changing environmental conditions requires a prospective inference of future actions and their consequences, a strategy also known as model-based decision making.1-3 In stable environments, extensive experience of actions and their consequences leads to a shift from a model-based to a model-free strategy, whereby behavioral selection is primarily governed by retrospective experiences of positive and negative outcomes. Human and animal studies, where subjects are required to speculate about implicit information and adjust behavioral responses over multiple sessions, point to a role for the central serotonergic system in model-based decision making.4-8 However, to directly test a causal relationship between serotonergic activity and model-based decision making, phase-specific manipulation of serotonergic activity is needed in a one-shot test, where learning by trial and error is neutralized. Moreover, the serotonergic origin responsible for this effect is yet to be determined. Herein, we demonstrate that optogenetic silencing of serotonin neurons in the dorsal raphe nucleus, but not in the median raphe nucleus, disrupts model-based decision making in lithium-induced outcome devaluation tasks.9-11 Our data indicate that the serotonergic behavioral effects are not due to increased locomotor activity, anxiolytic effects, or working memory deficits. Our findings provide insights into the neural mechanisms underlying neural weighting between model-free and model-based strategies.

Machine Learning-Based Decision Model to Distinguish Between COVID-19 and Influenza: A Retrospective, Two-Centered, Diagnostic Study.

BackgroundConsidering the current situation of the novel coronavirus disease (COVID-19) epidemic control, it is highly likely that COVID-19 and influenza may coincide during the approaching winter season. However, there is no available tool that can rapidly and precisely distinguish between these two diseases in the absence of laboratory evidence of specific pathogens.MethodsLaboratory-confirmed COVID-19 and influenza patients between December 1, 2019 and February 29, 2020, from Zhongnan Hospital of Wuhan University (ZHWU) and Wuhan No.1 Hospital (WNH) located in Wuhan, China, were included for analysis. A machine learning-based decision model was developed using the XGBoost algorithms.ResultsData of 357 COVID-19 and 1893 influenza patients from ZHWU were split into a training and a testing set in the ratio 7:3, while the dataset from WNH (308 COVID-19 and 312 influenza patients) was preserved for an external test. Model-based decision tree selected age, serum high-sensitivity C-reactive protein and circulating monocytes as meaningful indicators for classifying COVID-19 and influenza cases. In the training, testing and external sets, the model achieved good performance in identifying COVID-19 from influenza cases with a corresponding area under the receiver operating characteristic curve (AUC) of 0.94 (95% CI 0.93, 0.96), 0.93 (95% CI 0.90, 0.96), and 0.84 (95% CI: 0.81, 0.87), respectively.ConclusionMachine learning provides a tool that can rapidly and accurately distinguish between COVID-19 and influenza cases. This finding would be particularly useful in regions with massive co-occurrences of COVID-19 and influenza cases while limited resources for laboratory testing of specific pathogens.

Local and global spatio-temporal sensitivity analysis of thermal consolidation around a point heat source

Coupled thermo–hydro-mechanical (THM) models are used for the assessment of nuclear waste disposal, reservoir engineering, and many other branches of geotechnical engineering. Model-based decision making and optimization often entail sensitivity analyses (SA) and uncertainty quantification (UQ). The suitability of different UQ and SA methods for coupled THM problems on an engineering scale requires clarification. To provide such guidance without the need for large numerical studies, an analytical solution of a simple THM problem is employed here that encompasses the most relevant primary couplings, can robustly cover the entire parameter space, and remains computationally inexpensive. Both local (OVAT) and global sensitivity analysis (GSA) techniques are applied to the study of the THM model. The information that can be derived from the different approaches is then systematically assessed, such as parameters and interactions that control selected observation quantities and how their effect varies in both space and time. We provide application-oriented conclusions on the conditions which should be met when applying the different methods as well as examples for possible misinterpretations. The analysis can serve as a benchmark for UQ and SA software designed around numerical THM simulators. This work may also serve as an attempt to highlight the methodology and effectiveness of SA to the audience from an engineering background.

A Model-Based Decision Framework for the Multi-Depot Multi-Traveling Salesman Problem with Split and Delivery Demand considering different Key Performance Indicators

A Model-Based Decision Framework for the Multi-Depot Multi-Traveling Salesman Problem with Split and Delivery Demand considering different Key Performance Indicators

Serotonergic activity in the dorsal raphe nucleus is required for model-based decision making

Serotonergic activity in the dorsal raphe nucleus is required for model-based decision making

Implementation of machine learning model-based decision support system for healthcare professionals to predict T2DM risk using heart rate variability features

Non-invasive early diabetes prediction has been gaining much premarkable over the last decade. Heart rate variability (HRV) is the only non-invasive technique that can predict the future occurrence of the disease. Early prediction of diabetes can help doctors start an early intervention. To this end, the authors have developed a computational machine learning model to predict type 2 diabetes mellitus (T2DM) risk using heart rate variability features and have evaluated its robustness against the HRV of 50 patients data. The electrocardiogram (ECG) signal of the control population (n=40) and T2DM population (n=120) have been recorded in the supine position for 5 minutes, and HRV signals have been obtained. The time domain, frequency domain, and non-linear features have been extracted from the HRV signal. A decision support system has been developed based on a machine learning algorithm. Finally, the decision support system has been validated using the HRV features of 50 patients (Control n=10 and T2DM n=40). HRV features are selected for the prediction of T2DM. The decision support system has been designed using three machine learning models: Gradient boosting decision tree (GBDT), Extreme Gradient boosting (XGBoost), Categorical boosting (CatBoost), and their performance have been evaluated based on the Accuracy (ACC), Sensitivity (SEN), Specificity (SPC), Positive predicted value (PPV), Negative predicted value (NPV), False-positive rate (FPR), False-negative rate (FNR), F1 score, and Area under the receiver operating characteristic curve (AUC) metrics. The CatBoost model offers the best performance outcomes, and its results have been validated on 50 patients. Thus the CatBoost model can be use as a decision support system in hospitals to predict the risk of T2DM.

Digital Twins in Critical Care: What, When, How, Where, Why?

Abstract Healthcare and intensive care unit (ICU) medicine in particular, are facing a devastating tsunami of rising demand multiplied by increasing chronic disease and aging demographics, which is unmatched by society’s ability to pay. Digital technologies and automation have brought significant productivity gains to many industries, and manufacturing in particular, but not yet to medicine. In manufacturing, digital twins, model-based optimisation of manufacturing systems and equipment, are a rapidly growing means of further enhancing productivity and quality. This concept intersects well with the model-based decision support and control just beginning to emerge into clinical use, offering the opportunity to personalise care, and improve its quality and productivity. This article presents digital twins in a manufacturing concept and translates it into clinical practice, and then reviews the state of the art in key areas of ICU medicine.

PND43 Model-based Economic Evaluations of Interventions for Dementia: A Pragmatic Review

To identify and review new model-based economic evaluations of all types of interventions for people with dementia and their caregivers, since the systematic literature review (SLR) conducted by Nguyen et al. (2018). The search of PubMed and EMBASE databases conducted by Nguyen et al. (2018) was re-ran for the period 2018-present to identify model-based economic evaluations of interventions for dementia. Basic characteristics of each identified economic model (e.g. type of intervention, type of evaluation, perspective, population, modelling approach) were extracted for analysis and combined with the corresponding results from the original search. The search identified 263 records for screening, after removal of duplicates, and 20 publications were finally selected for inclusion, with 10 evaluating treatment with pharmaceuticals, 9 reporting prevention or diagnostic interventions and 1 covering non-pharmacological interventions. Only 4 out of the 20 studies were published after January 2018. The 16 additional studies were identified through citation chaining or full-text review of retrieved publications or excluded SLRs, and were included for completeness. 15 out of the 20 publications referred to interventions for Alzheimer’s Disease (AD). The majority of the evaluations use Markov models with a relatively simple representation of AD, commonly considering only cognitive function. One model-based economic evaluation (Green et al. 2019) identified in this search considers other features of AD such as functional and behavioral performances. Most of the studies reported taking a societal perspective. No new modelling methodologies were identified compared to those already reported by Nguyen et al. (2018). Nonetheless, an open-source model-based decision analytical framework has become available and is the only published model accounting for the key distinguishing AD symptoms domains (i.e. cognitive function, functional performance and behavioral and psychological symptoms), which may represent a shift in this research field.

Intensive Care Weaning (iCareWean) protocol on weaning from mechanical ventilation: a single-blinded multicentre randomised control trial comparing an open-loop decision support system and routine care, in the general intensive care unit

IntroductionAutomated systems for ventilator management to date have been either fully heuristic rule-based systems or based on a combination of simple physiological models and rules. These have been shown to...

Model-based decision making and model-free learning

Free will is anything but free. With it comes the onus of choice: not only what to do, but which inner voice to listen to - our 'automatic' response system, which some consider 'impulsive' or 'irrational', or our supposedly more rational deliberative one. Rather than a devil and angel sitting on our shoulders, research suggests that we have two decision-making systems residing in the brain, in our basal ganglia. Neither system is the devil and neither is irrational. They both have our best interests at heart and aim to suggest the best course of action calculated through rational algorithms. However, the algorithms they use are qualitatively different and do not always agree on which action is optimal. The rivalry between habitual, fast action and deliberative, purposeful action is an ongoing one.

Model-Based Decision Support Tools at Jugos S.A. Concentrated Fruit Juice Plant

We describe the development of a decision-support tool to assist in the operations of a large concentrated apple and pear juice plant. The tool’s objective is to generate detailed schedules of clarified juice batches to be produced in the following weeks considering incoming fruit forecasts, commercial commitments, and infrastructural constraints. The tool is based on two interactive modules, PLANNER and SIMOPT, with different and complementary purposes. Each module is based on mixed-integer models with specific inputs, outputs, and user interfaces. PLANNER consists of three submodules: (i) planning assigns a batch of concentrated juice to be produced on a specific day, taking into account cleaning activities, rest days, raw material availability, and production and storage constraints; (ii) preprocessing organizes juice orders in batches; and (iii) pooling provides a detailed monitoring of semielaborated juice in storage pools in terms of inventories and sugar and acid content. Finally, SIMOPT provides a detailed optimal operative condition of the plant together with a thorough calculation of specific costs. This information is used by PLANNER to evaluate the corresponding economic objective functions. Besides providing optimal target conditions to the plant and feasible production schedules, the developed tools generate production guidelines in the long term and allow performing scenario studies.

Policy Innovation and Governance for Irrigation Sustainability in the Arid, Saline San Joaquin River Basin

This paper provides a chronology and overview of events and policy initiatives aimed at addressing irrigation sustainability issues in the San Joaquin River Basin (SJRB) of California. Although the SJRB was selected in this case study, many of the same resource management issues are being played out in arid, agricultural regions around the world. The first part of this paper provides an introduction to some of the early issues impacting the expansion of irrigated agriculture primarily on the west side of the San Joaquin Valley and the policy and capital investments that were used to address salinity impairments to the use of the San Joaquin River (SJR) as an irrigation water supply. Irrigated agriculture requires large quantities of water if it is to be sustained, as well as supply water of adequate quality for the crop being grown. The second part of the paper addresses these supply issues and a period of excessive groundwater pumping that resulted in widespread land subsidence. A joint federal and state policy response that resulted in the facilities to import Delta water provided a remedy that lasted almost 50 years until the Sustainable Groundwater Management Act of 2014 was passed in the legislature to address a recurrence of the same issue. The paper describes the current state of basin-scale simulation modeling that many areas, including California, are using to craft a future sustainable groundwater resource management policy. The third section of the paper deals with unique water quality issues that arose in connection with the selenium crisis at Kesterson Reservoir and the significant threats to irrigation sustainability on the west side of the San Joaquin Valley that followed. The eventual policy response to this crisis was incremental, spanning two decades of University of California-led research programs focused on finding permanent solutions to the salt and selenium contamination problems constraining irrigated agriculture, primarily on the west side. Arid-zone agricultural drainage-induced water quality problems are becoming more ubiquitous worldwide. One policy approach that found traction in California is an innovative variant on the traditional Total Maximum Daily Load (TMDL) approach to salinity regulation, which has features in common with a scheme in Australia’s Hunter River Basin. The paper describes the real-time salinity management (RTSM) concept, which is geared to improving coordination of west side agricultural and wetland exports of salt load with east side tributary reservoir release flows to improve compliance with river salinity objectives. RTSM is a concept that requires access to continuous flow and electrical conductivity data from sensor networks located along the San Joaquin River and its major tributaries and a simulation model-based decision support designed to make salt load assimilative capacity forecasts. Web-based information dissemination and data sharing innovations are described with an emphasis on experience with stakeholder engagement and participation. The last decade has seen wide-scale, global deployment of similar technologies for enhancing irrigation agriculture productivity and protecting environmental resources.

Behavior-based scenario discovery using time series clustering

Scenario Discovery is a widely used method in model-based decision support for identifying common input space properties across ensembles of exploratory model runs. For model runs with behavior over time, these properties are identified by reducing each run to a single value, which obscures potentially decision-relevant dynamics. We address the problem of considering dynamics in Scenario Discovery by applying time series clustering to the ensemble of model runs, and then finding the common input properties for each cluster. This separates the input space into multiple scenarios, each corresponding to a distinct model dynamic. Policy interventions can be targeted at different scenarios by analyzing overlap of these subspaces. Our work expands Scenario Discovery by improving consideration of system behavior over time, which is highly relevant for the management of complex nonlinear systems such as ecosystems or technical infrastructure.

Optimized and Personalized Phlebotomy Schedules for Patients Suffering From Polycythemia Vera.

Polycythemia vera (PV) is a slow-growing type of blood cancer, where the production of red blood cells (RBCs) increase considerably. The principal treatment for targeting the symptoms of PV is bloodletting (phlebotomy) at regular intervals based on data derived from blood counts and physician assessments based on experience. Model-based decision support can help to identify optimal and individualized phlebotomy schedules to improve the treatment success and reduce the number of phlebotomies and thus negative side effects of the therapy. We present an extension of a simple compartment model of the production of RBCs in adults to capture patients suffering from PV. We analyze the model's properties to show the plausibility of its assumptions. We complement this with numerical results using exemplary PV patient data. The model is then used to simulate the dynamics of the disease and to compute optimal treatment plans. We discuss heuristics and solution approaches for different settings, which include constraints arising in real-world applications, where the scheduling of phlebotomies depends on appointments between patients and treating physicians. We expect that this research can support personalized clinical decisions in cases of PV.

Goal-oriented and habitual decisions: Neural signatures of model-based and model-free learning

Human decision-making is mainly driven by two fundamental learning processes: a slow, deliberative, goal-directed model-based process that maps out the potential outcomes of all options and a rapid habitual model-free process that enables reflexive repetition of previously successful choices. Although many model-informed neuroimaging studies have examined the neural correlates of model-based and model-free learning, the concordant activity among these two processes remains unclear. We used quantitative meta-analyses of functional magnetic resonance imaging experiments to identify the concordant activity pertaining to model-based and model-free learning over a range of reward-related paradigms. We found that: 1) both processes yielded concordant ventral striatum activity, 2) model-based learning activated the medial prefrontal cortex and orbital frontal cortex, and 3) model-free learning specifically activated the left globus pallidus and right caudate head. Our findings suggest that model-free and model-based decision making engage overlapping yet distinct neural regions. These stereotaxic maps improve our understanding of how deliberative goal-directed and reflexive habitual learning are implemented in the brain.

On the assimilation of environmental tracer observations for model-based decision support

Abstract. It has been advocated that history matching numerical models to a diverse range of observation data types, particularly including environmental tracer concentrations and their interpretations and derivatives (e.g., mean age), constitutes an effective and appropriate means to improve model forecast reliability. This study presents two regional-scale modeling case studies that directly and rigorously assess the value of discrete tritium concentration observations and tritium-derived mean residence time (MRT) estimates in two decision-support contexts; “value” is measured herein as both the improvement (or otherwise) in the reliability of forecasts through uncertainty variance reduction and bias minimization as a result of assimilating tritium or tritium-derived MRT observations. The first case study (Heretaunga Plains, New Zealand) utilizes a suite of steady-state and transient flow models and an advection-only particle-tracking model to evaluate the worth of tritium-derived MRT estimates relative to hydraulic potential, spring discharge and river–aquifer exchange flux observations. The worth of MRT observations is quantified in terms of the change in the uncertainty surrounding ecologically sensitive spring discharge forecasts via first-order second-moment (FOSM) analyses. The second case study (Hauraki Plains, New Zealand) employs paired simple–complex transient flow and transport models to evaluate the potential for assimilation-induced bias in simulated surface-water nitrate discharge to an ecologically sensitive estuary system; formal data assimilation of tritium observations is undertaken using an iterative ensemble smoother. The results of these case studies indicate that, for the decision-relevant forecasts considered, tritium observations are of variable benefit and may induce damaging bias in forecasts; these biases are a result of an imperfect model's inability to properly and directly assimilate the rich information content of the tritium observations. The findings of this study challenge the advocacy of the increasing use of tracers, and of diverse data types more generally, whenever environmental model data assimilation is undertaken with imperfect models. This study also highlights the need for improved imperfect-model data assimilation strategies. While these strategies will likely require increased model complexity (including advanced discretization, processes and parameterization) to allow for appropriate assimilation of rich and diverse data types that operate across a range of spatial and temporal scales commensurate with a forecast of management interest, it is critical that increased model complexity does not preclude the application of formal data assimilation and uncertainty quantification techniques due to model instability and excessive run times.

Model-oriented decision-making support system for the safe operation of the ship electric power plant improvement

A model-based decision support system is proposed to increase the reliability of the ship electric power plant. This system consists of a modeling unit and a unit for evaluating simulation results. The developed system can be implemented as part of control systems, both on the basis of programmable logic controllers, and on the basis of a separate programmable logic controller. The mathematical model of the ship electric power installation is given. The criteria used to evaluate the results of modeling and possible consequences are considered. A computer program for the implementation of this system presented. The results of modeling various emergency situations are given. Considered the possibility of using this system for training.

Uncertainty Quantification and Propagation in Computational Materials Science and Simulation-Assisted Materials Design

Significant advances in theory, simulation tools, advanced computing infrastructure, and experimental frameworks have enabled the field of materials science to become increasingly reliant on computer simulations. Theory-grounded computational models provide a better understanding of observed materials phenomena. At the same time, computational tools constitute an important ingredient of any framework that seeks to accelerate the materials development cycle. While simulations keep increasing in sophistication, formal frameworks for the quantification, propagation, and management of their uncertainties are required. Uncertainty analysis is fundamental to any effort to validate and verify simulations, which is often overlooked. Likewise, no simulation-driven materials design effort can be done with any level of robustness without properly accounting for the uncertainty in the predictions derived from the computational models. Here, we review some of the most recent works that have focused on the analysis, quantification, propagation, and management of uncertainty in computational materials science and ICME-based simulation-assisted materials design. Modern concepts of efficient uncertainty quantification and propagation, multi-scale/multi-level uncertainty analysis, model selection as well as model fusion are also discussed. While the topic remains relatively unexplored, there have been significant advances that herald an increased sophistication in the approaches followed for model validation and verification and model-based decision support.

A framework for data-driven design in a product innovation process: data analysis and visualisation for model-based decision making

The paper presents a four-layer framework for the application of data-driven design in a product innovation process. The framework builds on the Knowledge Value Stream and on the Product Value Streams of a product innovation process and indicates how data-driven activities shall be structured and organised in relation to the different phases of a model-based decision process. Visualisation is proposed as a communication enabler at the top of the framework to overcome the comprehensibility barrier between data science and engineering design models. The framework is implemented in the case study of a construction equipment encompassing the analysis of operational machine data and the experimentation of suitable visualisation techniques. Ultimately, a list of challenges for the implementation of data-driven design is presented, and the capability of the framework to support the transition toward data-driven design is discussed in relation to the emergence of product-service systems solutions.

A framework for data-driven design in a product innovation process: data analysis and visualisation for model-based decision making

The paper presents a four-layer framework for the application of data-driven design in a product innovation process. The framework builds on the Knowledge Value Stream and on the Product Value Streams of a product innovation process and indicates how data-driven activities shall be structured and organised in relation to the different phases of a model-based decision process. Visualisation is proposed as a communication enabler at the top of the framework to overcome the comprehensibility barrier between data science and engineering design models. The framework is implemented in the case study of a construction equipment encompassing the analysis of operational machine data and the experimentation of suitable visualisation techniques. Ultimately, a list of challenges for the implementation of data-driven design is presented, and the capability of the framework to support the transition toward data-driven design is discussed in relation to the emergence of product-service systems solutions.