Distributed parallel compilation of MSBNs

Multiply sectioned Bayesian networks (MSBNs) support multiagent probabilistic inference in distributed large problem domains. Inference with MSBNs can be performed using their compiled representations. The compilation involves moralization and triangulation of a set of local graphical structures. Privacy of agents may prevent us from compiling MSBNs at a central location. In earlier work, agents performed compilation sequentially via a depth‐first traversal of the hypertree that organizes local subnets, where communication failure between any two agents would crush the whole work. In this paper, we present an asynchronous compilation method by which multiple agents compile MSBNs in full parallel. Compared with the traversal compilation, the asynchronous one is robust, self‐adaptive, and fault‐tolerant. Experiments show that both methods provide similar quality compilation to simple MSBNs, but the asynchronous one provides much higher quality compilation to complex MSBNs. Empirical study also indicates that the asynchronous one is consistently faster than the traversal one. Copyright © 2009 John Wiley & Sons, Ltd.     

Using OWL ontologies for adaptive patient information modelling and preoperative clinical decision support

We here present our research and experience regarding the design and implementation of a knowledge-based preoperative assessment decision support system. We discuss generic design considerations as well as the practical system implementation. We developed the system using semantic web technology, including modular ontologies developed in the OWL web ontology language, the OWL Java application programming interface and an automated logic reasoner. We discuss how the system enables to tailor patient information collection according to personalized medical context. The use of ontologies at the core of the system’s architecture permits to efficiently manage a vast repository of preoperative assessment domain knowledge, including classification of surgical procedures, classification of morbidities and guidelines for routine preoperative tests. Logical inference on the domain knowledge according to individual patient’s medical context enables personalized patients’ reports consisting of a risk assessment and clinical recommendations such as relevant preoperative tests.     

Support for real-time decision making in mobile financial applications

Mobile users making real-time decisions based on current information need confidence that their context has been taken into consideration in producing the system’s recommendations. This chapter reviews current use of mobile technologies for context-aware real-time decision support. Specifically, it describes a framework for assessing the impact of mobility in decision making. The framework uses dynamic context model of data quality to represent uncertainties in the mobile decision-making environment. This framework can be used for developing visual interactive displays for communicating to the user relevant changes in data quality when working in mobile environments. As an illustration, this chapter proposes a real-time decision support procedure for on-the-spot assistance to the mobile consumer when choosing the best payment option to efficiently manage their budget. The proposed procedure is based on multi-attribute decision analysis, scenario reasoning, and a quality of data framework. The feasibility of the approach is demonstrated with a mobile decision-support system prototype implementation. This article is part of the “Handbook on Decision Support Systems” edited by Frada Burstein and Clyde W. Holsapple (2008) Springer.     

Optimal selection and placement of BMPs and LID practices with a rainfall-runoff model

A decision support tool, which links a hydrologic/water quality model (L-THIA-LID 2.1) with optimization algorithms (AMALGAM) using a computational efficiency framework (MLSOPT), was developed to optimally implement BMPs and LID practices to reduce runoff and pollutant loads. The decision support tool was applied in the Crooked Creek watershed, Indiana, USA. For initial expenditures on practices, the environmental benefits increased rapidly as expenditures increased. However, beyond certain expenditure levels, additional spending did not result in noticeable additional environmental impacts. Compared to random placement of practices, the optimization strategy provided 3.9–7.7 times the level of runoff/pollutant load reductions for the same expenditures. To obtain the same environmental benefits, costs of random practices placement were 4.2–14.5 times the optimized practice placement cost. The decision support tool is capable of supporting decision makers in optimally selecting and placing BMPs and LID practices to obtain maximum environmental benefits with minimum costs.     

How do Experts Adapt their Explanations to a Layperson’s Knowledge in Asynchronous Communication? An Experimental Study

Despite a plethora of recommendations for personalization techniques, such approaches often lack empirical justification and their benefits to users remain obscure. The study described in this paper takes a step towards filling this gap by introducing an evidence-based approach for deriving adaptive interaction techniques. In a dialogue experiment with 36 dyads of computer experts and laypersons, we observed how experts tailored their written explanations to laypersons’ communicational needs. To support adaptation, the experts in the experimental condition were provided with information about the layperson’s knowledge level. In the control condition, the experts had no available information. During the composition of their answers, the experts in both conditions articulated their planning activities. Compared with the control condition, the experts in the experimental condition made a greater attempt to form a mental model about the layperson’s knowledge. As a result, they varied the type and proportion of the information they provided depending on the layperson’s individual knowledge level. Accordingly, such adaptive explanations helped laypersons reduce comprehension breakdowns and acquire new knowledge. These results provide evidence for theoretical assumptions regarding cognitive processes in text production and conversation. They empirically ground and advance techniques for adaptation of content in adaptive hypermedia systems. They are suggestive of ways in which explanations in recommender and decision support systems could be effectively adapted to the user’s knowledge background and goals.     

Case studies of IT sophistication in nursing homes: A mixed method approach to examine communication strategies about pressure ulcer prevention practices

Most nursing homes lack information technology (IT) for supporting clinical work in spite of its potential to improve the safety, quality, and efficiency of nursing home care in the United States. Increased attention to medical error and concern for patient safety have prompted general recommendations to develop sophisticated technologies to support clinical decision making at the point of care, to promote data standards in electronic records, and to develop systems that communicate with each other. However, little is known about what IT applications best support communication and risk assessment practices to improve resident outcomes in nursing homes. Thus, the overall aim of this study was to evaluate how differences in IT sophistication in nursing homes impact communication and use of technology related to skin care and pressure ulcers. We used a mixed method approach to conduct case studies on two nursing homes – one with high IT sophistication and one with low IT sophistication. Observational analysis and social network analysis were used to identify patterns in communication types and locations; also, focus groups were conducted to explore communication strategies used by Certified Nursing Assistants (CNAs) to support pressure ulcer prevention practices. Overall, results from social network analysis of observational data indicate that direct interactions between CNAs and registered nurses (RNs) or licensed practical nurses (LPNs) were more frequent in the low IT sophistication home and occurred in more centralized locations (e.g. the nursing station) compared to the high IT sophistication home. Moreover, these findings are supported by focus group results, which indicate that the high IT sophistication home had more robust and integrated communication strategies (both IT and non IT) that may allow for interactions throughout the facility and require less frequent face to face interactions between CNAs and RNs or LPNs to verify orders or report patient status. Results from this study provide insight into the design and assessment of different forms of communication to support clinical work in NHs.Relevance to industryNurses bear great burdens for nursing home care; yet, issues persist with poor quality, variable performance of caregiving, and lack of implementation of proven care interventions. One new hope for improvement in nursing home care is the introduction of IT to improve communication, clinical decision-making, and quality of care.     

A systematic mapping study on best practices for domain-specific modeling

Model-driven software development comes in different styles. While standard-based approaches leverage existing language standards (e.g., UML), tooling, and development processes, domain-specific modeling (DSM) requires languages and tool support to be created prior to the actual software development. The design, implementation, and testing of languages and tool support require a wide spectrum of methods and techniques which introduce complexity and new challenges. To tackle these DSM-specific challenges, best practices have been collected from various application domains and published in literature to guide the development and application of DSM solutions. It is the goal of this paper to identify studies reporting best practices on domain-specific modeling. Moreover, a systematic and comprehensive compilation of best practices should act as a starting point to identify literature that facilitates industrial adoption of DSM. To search for literature and classify identified studies, we conduct a systematic mapping study (SMS). Furthermore, we perform an in-depth analysis of the identified studies to answer how practices overlap, complement, or contradict each other. The systematic search resulted in 21 studies reporting 321 best practices. From these 321 practices, we compiled 192 unique best practices. We found that the DSM community created a substantial corpus of best practices for DSM. The large majority of practices (75%) are only reported once. The top best practices by number of reports contain only practices that are reported at least four times. However, the frequency of reports does not necessarily imply a high importance of practice, as the application of practices is always context specific.     

Factors affecting cross-hospital exchange of Electronic Medical Records

Adopting Electronic Medical Record (EMR) exchange may save patients’ lives and also improve their health. Most Electronic Data Interchange studies have considered commercial enterprises and necessary funding to support business activities, EMR exchange focuses on saving patients’ lives. Our study combined transaction cost and sociological perspectives to identify factors that affect a hospitals’ willingness to implement EMR exchange. A survey was conducted with regional hospitals and medical centers in Taiwan to justify the validity of a resulting model. Our findings indicated that the model was valuable and determined which factors influenced the decision to implement EMR exchange at these hospitals (perceived benefits, uncertainty, influence, and reciprocal investments). Based on these findings, healthcare policy makers can promote EMR exchange and hospitals can identify desirable partners that will form a strategic alliance to meet the dynamic challenges in the healthcare industry.     

Towards best practice implementation and application of models for analysis of water resources management scenarios

Water resources management models are widely used to evaluate planning or operational scenarios to support water resource management decision-making. However, the approaches to modelling used in the past have led to problems, such as modellers having difficulty establishing the credibility of their model with stakeholders, and stakeholders having difficulty understanding and trusting model results. A best practice approach to the implementation and application of water resources management models based on a quality assurance procedure is an appropriate means of overcoming these difficulties, and there are a number of guidelines and papers available promoting this approach. However, guidance in these on the use of models to analyse water resource planning scenarios is limited or not provided. This paper therefore provides guidance on the implementation and application of water resources management models with an emphasis on scenario analysis. This guidance is principally intended for practising modellers, and also for peer reviewers and stakeholders such as managers, decision makers, and community-based groups. Adoption strategies and recommendations for future directions are also discussed.     

Model use in sustainability policy making: An experimental study

We examine whether decision makers who use a system dynamics model more readily create policies whose outcomes match their stated priorities or are on the Pareto Frontier of achieved outcomes, and whether model users are more likely to change their priorities. Comparing model use with other decision tools, we address these questions in a role-play simulation (serious game) experiment in which participants make multi-dimensional policy recommendations to address environmental, economic, and social equity dimensions of sustainability. Participants are randomly assigned to use one of four decision support tools: (1) actively using the model or passively being briefed. The briefings cover: (2) the insights of the model; (3) general, relevant information; or (4) irrelevant information (control). We find that model-users, group (1), discovered key insights about win-win opportunities—over-performing their priorities—and most readily created policies on the Pareto Frontier of achieved outcomes. Groups (2) and (3) best matched their policy outcomes to their priorities, missing the win-win.     

Evidence-based recommendations for building better bras for women treated for breast cancer

Participating in exercise is beneficial for women who have been treated for breast cancer. However, not being able to find a comfortable exercise bra can be a barrier to exercise participation. This study aimed to systematically investigate what breast support women treated for breast cancer want when they exercise in order to provide evidence-based recommendations to improve exercise bra designs for these women. Based on 432 responses from a national online survey, frequency and relationship data were analysed (binary logistic regression) to understand exercise bra issues pertinent to this population. These issues included being able to control for asymmetrical cup sizes, managing heightened skin sensitivity, managing fluid (size) fluctuations, managing a prosthesis and restoring body image by restoring shape. This study provides evidence-based recommendations to inform an exercise bra design that will meet the unique needs of women treated for breast cancer. Rigorous, evidence-based evaluations of exercise bras for women treated for breast cancer may contribute to their well-being and quality of life through enhanced designs.     

Crisis communication,learning and responding: Best practices in social media

As noted by Seeger (2006) the notion of best practices is often use to improve professional practice; to create research and functional recommendations to use in a specific situation. This essay describes best practices in crisis communication specifically through the use of social media. It provides suggestions and approaches for improving the effectiveness of crisis communication and learning with and between organizations, governments and citizens. Seven best practices for effective crisis communication using social media are outlined.     

A decision support system for paper making based on simulation and optimization

In this paper a model-driven decision support system related to paper making is introduced. The intention is to emphasize the necessity of coupling different modeling techniques, multiobjective optimization, and software engineering in order to make the end user application realistic, practical and usable. Firstly the paper making process and selected aspects concerning its mathematical modeling, numerical simulation, and multiobjective optimization are introduced, then the related computerized system, called a virtual paper making line, is described. In addition, the associated decision support system, which provides a suitable level of automation to improve the quality of decision making and support the user’s expertise is discussed. Finally, an example presents different ways of using such a software tool.     

From what if to what's best in DSS

The author of IFPS/OPTIMUM presents some typical applications of optimization within decision support systems (DSS). They show that finding ‘what's best’ is gradually superseding the ‘what if’ and ‘goal seeking’ type of analysis that is prevalent now. IFPS/OPTIMUM is now used by over 130 companies, including many of the Fortune 500 ones. The main drawback of ‘what if’ and ‘goal seeking’ type of analysis is that one cannot impose restrictions on model variables and have the DSS generator solve for values of the ‘what if’ or ‘goal seek’ variables that satisfy those restrictions. Instead, one would have to muddle through many ‘what if's’ or ‘goal seeks’ at random and check if the restrictions are satisfied. It is a manual process that is very time consuming. ‘What's best’ analysis with optimization allows one to impose such restrictions easily and do the analysis quickly. This type of analysis is playing an increasingly significant role within decision support systems. Practicing management scientists and managers should note this phenomenon. New tools and new ways of addressing problems are emerging.     

Development and maintenance of guideline-based decision support for pharmacological treatment of hypertension

The objective was to build a computer-based decision support system (DSS), which could apply the formal rules embedded in guidelines regarding pharmacological treatment of hypertension. The aim was also to test VISUAL BASIC as a development tool for DSS's in health care. From the Swedish guidelines for treatment of hypertension, the most widely accepted and scientifically best proved treatment strategies were chosen and implemented as rules. A DSS that is capable of applying the evidence-based rules extracted from guidelines regarding drug treatment of hypertension, to any patient's medical profile, was constructed. The output consists of a recommendation regarding preferred generic drug class and also a written report, reflecting decision steps provided by the rule-base and inference engine. We also provide methods for formalising an implementable language of guidelines. A mainstream programming language like VISUAL BASIC can be an alternative when building complicated decision support systems. A logic formal notation can facilitate communication between the expert and the programmer. The program is a stand-alone product independent of computerized medical records and thereby easy to install and maintain.     

Practice, systems and technology for seniors

This paper explores how research teams in Intel’s Digital Health Group are using ethnography to identify ‘designable moments’—spaces, times, objects, issues and practices which suggest opportunities for appropriate interventions. It argues that technology innovation should aim to incorporate the views, experiences and practices of users from the start of the design process to support independent living and develop culturally sensitive enhancements that contribute towards wellbeing and a life of quality for local older populations.