Rule-based Decision Support System Research Articles

Architecture and Development Framework for a Web-Based Risk Assessment and Management Platform Developed on WordPress to Address Opioid Overdose.

The number of overdose-related fatalities continues to reach historic levels across Canada, despite ongoing efforts by authorities. To reduce mortality, a clinical trajectory ranging from preventative measures to crisis intervention, skill training to treatment, and risk assessment to risk management needs to be supported. The web-based Risk Assessment and Management Platform (RAMP) was developed to realize this concept and to empower people who use drugs through an integrated tool that allows them to better understand and manage their risk of overdose. This paper outlines the architecture and development of RAMP, which is built on the WordPress platform. WordPress components are mapped onto a 3-tier architecture that consists of presentation, application, and database layers. The architecture facilitates the development of a modular software that includes several features that are independent in functionality but interact with each other in an integrated platform. The relatively low coupling and high coherence of the features may reduce the cost of maintenance and increase flexibility of future developments. RAMP's architecture comprises a user interface, conceptual framework, and backend layers. The RAMP front end effectively uses some of the WordPress' features such as HTML5, CSS, and JavaScript to create a mobile, friendly, and scalable user interface. The RAMP backend uses several standard and custom WordPress plug-ins to support risk assessment and monitoring, with the goal of mitigating the impacts and eliminating risks together. A rule-based decision support system has been hard-coded to suggest relevant modules and goals to complement each user's lifestyle and goals based on their risk assessment. Finally, the backend uses the MySQL database management system and communicates with the RAMP framework layer via the data access layer to facilitate a timely and secure handling of information. Overall, RAMP is a modular system developed to identify and manage the risk of opioid overdose in the population of people who use drugs. Its modular design uses the WordPress architecture to efficiently communicate between layers and provide a base for external plug-ins. There is potential for the current system to adopt and address other related fields such as suicide, anxiety, and trauma. Broader implementation will support this concept and lead to the next level of functionality.

Golden Search Optimization based adaptive and diagonal kernel convolution neural network for disease prediction and securing IoT data in cloud

Due to their open architecture, Internet of Things (IoT) devices have limited user access and are susceptible to various security vulnerabilities. When healthcare data is stored in the cloud, it often becomes vulnerable to side-channel attacks, Denial of Service (DoS) attacks, and eavesdropping. Consequently, there is a need for an efficient access control scheme to enhance the data-sharing processes of IoT devices. Different machine learning and rule-based decision support systems have been proposed to improve the security of cloud-based data storage. However, these models have exhibited increased false alarm rates, complex processing, and longer response times. To address this research gap, we propose a Fuzzy TOPSIS model based on Golden Search Optimization (GSO) to enhance the security of patient information at both IoT and cloud server levels. Additionally, we introduce a revolutionary Diagonal Kernel Convolution Neural Network with an adaptable kernel (DKCNN-AK) to enhance the accuracy of lung disease diagnosis. The GSO algorithm categorizes the security ranks determined by the fuzzy TOPSIS model into the best and worst security principles. We conducted simulations using the COVID-19 X-Ray detection dataset, the lung cancer dataset, and the data collected from IoT devices for analysis.The experiments were conducted to assess various aspects, including computational time, security level analysis, information leakage, privacy level, prediction accuracy, sensitivity, specificity, and confusion matrix. The improved prediction accuracy and security provided by our proposed approach demonstrate its efficiency

Factors Hindering Solar Photovoltaic System Implementation in Buildings and Infrastructure Projects: Analysis through a Multiple Linear Regression Model and Rule-Based Decision Support System

Energy plays a predominant role in the development of society. With advancements in technology and the growth of society (buildings and infrastructures), the demand for energy is rapidly increasing. Developing countries typically rely on the import of fossil fuels and capital investments in infrastructure development to meet their energy needs. The execution of solar PV projects in developing countries is currently not being implemented promisingly. Therefore, the determination of the critical success factors hindering the implementation of solar PV projects is the need of the hour. The aim of this study is to determine the factors that hinder the implementation of solar PV projects through the use of a multiple linear regression model (MLRM) and a rule-based decision support system (RBDSS). Seven categories of factors were identified through a detailed literature review and interviews with energy experts. Four hundred and twenty-nine complete responses were collected in total through a questionnaire, and they were analyzed using relative importance indexing (RII) and MLRM and RBDSS approaches. A comparison was carried out against both methodologies to determine the most critical barriers to the implementation of solar PV projects. The findings regarding the MLRM approach showed that the top seven critical factors were economic conditions, encouraging policies, technological knowledge, organizational support, social awareness, market stability, and miscellaneous aspects. Similarly, the results for the RBDSS approach identified that the top seven critical factors were encouraging policies, economic conditions, organizational support, technological knowledge, market stability, social awareness, and miscellaneous aspects. The application of MLRM and RBDSS will help stakeholders in making timely decisions and corrections during the implementation phase, providing a systematic way to support the performance and execution of solar projects.

An Integrated Monitoring System for Coastal and Riparian Areas Based on Remote Sensing and Machine Learning

Nowadays, coastal areas are exposed to multiple hazards of increasing severity, such as coastal floods, erosion, subsidence due to a combination of natural and anthropogenic factors, including climate change and urbanisation. In order to cope with these challenges, new remote sensing monitoring solutions are required that are based on knowledge extraction and state of the art machine learning solutions that provide insights into the related physical mechanisms and allow the creation of innovative Decision Support Tools for managing authorities. In this paper, a novel user-friendly monitoring system is presented, based on state-of-the-art remote sensing and machine learning approaches. It uses processes for collecting and analysing data from various heterogeneous sources (satellite, in-situ, and other auxiliary data) for monitoring land cover and land use changes, coastline changes soil erosion, land deformations, and sea/ground water level. A rule-based Decision Support System (DSS) will be developed to evaluate changes over time and create alerts when needed. Finally, a WebGIS interface allows end-users to access and visualize information from the system. Experimental results deriving from various datasets are provided to assess the performance of the proposed system, which is implemented within the EPIPELAGIC bilateral Greece-China project. The system is currently being installed in the Greek case study area, namely Thermaikos Gulf in Thessaloniki, Greece.

Personalized Recommendations for Physical Activity e-Coaching (OntoRecoModel): Ontological Modeling.

BackgroundAutomatic e-coaching may motivate individuals to lead a healthy lifestyle with early health risk prediction, personalized recommendation generation, and goal evaluation. Multiple studies have reported on uninterrupted and automatic monitoring of behavioral aspects (such as sedentary time, amount, and type of physical activity); however, e-coaching and personalized feedback techniques are still in a nascent stage. Current intelligent coaching strategies are mostly based on the handcrafted string messages that rarely individualize to each user’s needs, context, and preferences. Therefore, more realistic, flexible, practical, sophisticated, and engaging strategies are needed to model personalized recommendations.ObjectiveThis study aims to design and develop an ontology to model personalized recommendation message intent, components (such as suggestion, feedback, argument, and follow-ups), and contents (such as spatial and temporal content and objects relevant to perform the recommended activities). A reasoning technique will help to discover implied knowledge from the proposed ontology. Furthermore, recommendation messages can be classified into different categories in the proposed ontology.MethodsThe ontology was created using Protégé (version 5.5.0) open-source software. We used the Java-based Jena Framework (version 3.16) to build a semantic web application as a proof of concept, which included Resource Description Framework application programming interface, World Wide Web Consortium Web Ontology Language application programming interface, native tuple database, and SPARQL Protocol and Resource Description Framework Query Language query engine. The HermiT (version 1.4.3.x) ontology reasoner available in Protégé 5.x implemented the logical and structural consistency of the proposed ontology. To verify the proposed ontology model, we simulated data for 8 test cases. The personalized recommendation messages were generated based on the processing of personal activity data in combination with contextual weather data and personal preference data. The developed ontology was processed using a query engine against a rule base to generate personalized recommendations.ResultsThe proposed ontology was implemented in automatic activity coaching to generate and deliver meaningful, personalized lifestyle recommendations. The ontology can be visualized using OWLViz and OntoGraf. In addition, we developed an ontology verification module that behaves similar to a rule-based decision support system to analyze the generation and delivery of personalized recommendation messages following a logical structure.ConclusionsThis study led to the creation of a meaningful ontology to generate and model personalized recommendation messages for physical activity coaching.

Refined Segmentation of K-Means Cluster Using Hidden-Markov Random Field (HMRF) for Leukemia Detection

Abstract: An important part in detecting and diagnosing leukaemia involves the microscopic examination of a patient’s blood sample. The type of malignant cells called the blasts is frequently determined by morphological alterations in White Blood Cells (WBCs). Real-time procedures are time-consuming, error-prone, and expensive. In order to overcome this problem the features such as the colour, shapes and texture can be extracted from the original image. This paper deals with the segmentation techniques. The segmentation process in automated approach is developed using k-means clustering (KMC) and Hidden-Markov Random Field (HMRF). This is also termed as Refined Segmentation (Re-Se). This analysis includes Aspirate Smear Images (ASI). First cell segmentation using a KMC is performed, then a Cell Image Representing Model (CIRM) using HMRF is build, estimating prototype variables using Expectation Maximization (EM) ratio, iterating until the best value is found, and finally achieving next phase called Refined Segmentation (Re-Se). Then, features including texture, geometry, colour, and statistical information are retrieved and used to categorize leukaemia cells from WBCs, by using Fuzzy Rule-Based Decision Support System (FRDSS). By incorporating more feature extraction techniques, the suggested method improves a blast of most immature cells identification for the Acute Myeloid Leukemia (AML) detection

A Digital Advisor Twin for Crop Nitrogen Management

Farmers and consultants face an unmanageable amount of diverse knowledge and information for crop management decisions. To determine optimal actions, decision makers require knowledge-based support. In this way, decisions can be improved and heuristics can be replaced over time. The study presents a digital knowledge base with an integrated decision support system (DSS), using the example of nutrient supply, specifically nitrogen (N), fertilization. Therefore, the requirements of farmers and crop consultants for DSS to inform fertilization decisions for winter wheat (Triticum aestivum L.) were elaborated using surveys, expert interviews, and a prototype test. Semantic knowledge was enriched by expert knowledge and combined in a web application, the Crop Portal. To map regional and personal decision making patterns and experiences, the tacit knowledge on the complex advisory problem of N fertilization is made digitally usable. For this purpose, 16 fuzzy variables were specified and formalized. Individual decision trees and their interactions with an integrative knowledge base were used to multiply the consulting reach of experts. Using three consultants and nine model farms from different soil–climate areas in Germany, the Crop Portal was tested under practical conditions and the perceived pragmatic and hedonic quality of the system was evaluated using a standardized questionnaire. The field test showed that the variation in fertilizer recommendations from the ‘digital advisor twin’ ranged from 5 kg N ha−1 to 16 kg N ha−1 when compared with the decisions of the experts in the field. The study presents the participatory development and evaluation of a rule-based DSS prototype in agricultural practice.

A rule-based decision support system for aiding iron deficiency management

Iron is a vital mineral for the proper function of hemoglobin which is also a protein needed to transport oxygen in the blood. The lack of iron in human blood causes a range of serious health problems including "anemia." In this article, the COntAneRS (Clinical ONTology-based Iron Deficiency-ANEmia- Recommendation System) is proposed as a clinical decision support system to diagnose iron deficiency and manage its treatment. The applied methodologies and main technical contributions of this study are discussed in four aspects: (1) Iron Deficiency Domain Ontology (IDDOnt), (2) Semantic Web Rule Knowledgebase (SWRL), (3) Inference Engine, and (4) Physician Portal of the system. Experimental studies of the proposed system have been applied on a population of 200 people, consisting of real anemia patients and healthy individuals. First, a decision tree classifier is used to diagnose iron deficiency condition based on the patients' demographic information and certain medical data, as well as recently measured hemoglobin CBC levels of the patients. To check the effectiveness of the system, the data of 50 anonymous patients randomly selected from 200 patients are entered manually in the IDDOnt and the system is then verified according to the inferencing results. After inferencing step, the recommendations related to appropriate iron drugs, daily consumption dose, drug consumption periods, and additional medical suggestions about drug interactions are provided by the system to the responsible physician through system ontology, SWRL rules, and web services. As a result of experimental studies, our system has provided very good accuracy (99.5%) and robust results in producing patient-suitable suggestions. In addition, the applicability of the system on the cases is discussed as case studies in this paper. The results reported from the applied case studies are promising in demonstrating the applicability, effectiveness, and efficiency of the proposed approach.

Change Point Enhanced Anomaly Detection for IoT Time Series Data

Due to the exponential growth of the Internet of Things networks and the massive amount of time series data collected from these networks, it is essential to apply efficient methods for Big Data analysis in order to extract meaningful information and statistics. Anomaly detection is an important part of time series analysis, improving the quality of further analysis, such as prediction and forecasting. Thus, detecting sudden change points with normal behavior and using them to discriminate between abnormal behavior, i.e., outliers, is a crucial step used to minimize the false positive rate and to build accurate machine learning models for prediction and forecasting. In this paper, we propose a rule-based decision system that enhances anomaly detection in multivariate time series using change point detection. Our architecture uses a pipeline that automatically manages to detect real anomalies and remove the false positives introduced by change points. We employ both traditional and deep learning unsupervised algorithms, in total, five anomaly detection and five change point detection algorithms. Additionally, we propose a new confidence metric based on the support for a time series point to be an anomaly and the support for the same point to be a change point. In our experiments, we use a large real-world dataset containing multivariate time series about water consumption collected from smart meters. As an evaluation metric, we use Mean Absolute Error (MAE). The low MAE values show that the algorithms accurately determine anomalies and change points. The experimental results strengthen our assumption that anomaly detection can be improved by determining and removing change points as well as validates the correctness of our proposed rules in real-world scenarios. Furthermore, the proposed rule-based decision support systems enable users to make informed decisions regarding the status of the water distribution network and perform effectively predictive and proactive maintenance.

Use of Decision Trees for the Development of Decision Support Systems for the Control of Grinding Circuits

Grinding circuits can exhibit strong nonlinear behaviour, which may make automatic supervisory control difficult and, as a result, operators still play an important role in the control of many of these circuits. Since the experience among operators may be highly variable, control of grinding circuits may not be optimal and could benefit from automated decision support. This could be based on heuristics from process experts, but increasingly could also be derived from plant data. In this paper, the latter approach, based on the use of decision trees to develop rule-based decision support systems, is considered. The focus is on compact, easy to understand rules that are well supported by the data. The approach is demonstrated by means of an industrial case study. In the case study, the decision trees were not only able to capture operational heuristics in a compact intelligible format, but were also able to identify the most influential variables as reliably as more sophisticated models, such as random forests.

An Automatic Ontology-Based Approach to Support Logical Representation of Observable and Measurable Data for Healthy Lifestyle Management: Proof-of-Concept Study.

BackgroundLifestyle diseases, because of adverse health behavior, are the foremost cause of death worldwide. An eCoach system may encourage individuals to lead a healthy lifestyle with early health risk prediction, personalized recommendation generation, and goal evaluation. Such an eCoach system needs to collect and transform distributed heterogenous health and wellness data into meaningful information to train an artificially intelligent health risk prediction model. However, it may produce a data compatibility dilemma. Our proposed eHealth ontology can increase interoperability between different heterogeneous networks, provide situation awareness, help in data integration, and discover inferred knowledge. This “proof-of-concept” study will help sensor, questionnaire, and interview data to be more organized for health risk prediction and personalized recommendation generation targeting obesity as a study case.ObjectiveThe aim of this study is to develop an OWL-based ontology (UiA eHealth Ontology/UiAeHo) model to annotate personal, physiological, behavioral, and contextual data from heterogeneous sources (sensor, questionnaire, and interview), followed by structuring and standardizing of diverse descriptions to generate meaningful, practical, personalized, and contextual lifestyle recommendations based on the defined rules.MethodsWe have developed a simulator to collect dummy personal, physiological, behavioral, and contextual data related to artificial participants involved in health monitoring. We have integrated the concepts of “Semantic Sensor Network Ontology” and “Systematized Nomenclature of Medicine—Clinical Terms” to develop our proposed eHealth ontology. The ontology has been created using Protégé (version 5.x). We have used the Java-based “Jena Framework” (version 3.16) for building a semantic web application that includes resource description framework (RDF) application programming interface (API), OWL API, native tuple store (tuple database), and the SPARQL (Simple Protocol and RDF Query Language) query engine. The logical and structural consistency of the proposed ontology has been evaluated with the “HermiT 1.4.3.x” ontology reasoner available in Protégé 5.x.ResultsThe proposed ontology has been implemented for the study case “obesity.” However, it can be extended further to other lifestyle diseases. “UiA eHealth Ontology” has been constructed using logical axioms, declaration axioms, classes, object properties, and data properties. The ontology can be visualized with “Owl Viz,” and the formal representation has been used to infer a participant’s health status using the “HermiT” reasoner. We have also developed a module for ontology verification that behaves like a rule-based decision support system to predict the probability for health risk, based on the evaluation of the results obtained from SPARQL queries. Furthermore, we discussed the potential lifestyle recommendation generation plan against adverse behavioral risks.ConclusionsThis study has led to the creation of a meaningful, context-specific ontology to model massive, unintuitive, raw, unstructured observations for health and wellness data (eg, sensors, interviews, questionnaires) and to annotate them with semantic metadata to create a compact, intelligible abstraction for health risk predictions for individualized recommendation generation.

Covid-19 Series: A rule-based decision support system for analysis behavior of people while working from home

COVID-19 makes the community must carry out activities such as school, work, and worship at home. However, the long-running activities from home make people experience boredom which can lead to stress. On the other side, the entertainment obtained by the public through smartphones by reading articles by their interests can reduce boredom. This paper proposed a rules-based decision support system to help the people to make choices of their activities from home while COVID-19 e rules-based approach for make an application decision support system. Ruled base used in application to selected process through characteristics following the interests of the community. This decision support system is implemented in mobile web applications. The system can display articles based on interests by questions or statements through the front end system. The results showed that most users of the application in a happy condition while working from home.

Evidential Reasoning Rule-Based Decision Support System for Predicting ICU Admission and In-Hospital Death of Trauma

We propose to employ evidential reasoning (ER) rule to construct a clinical decision support system (CDSS) to aid physicians to predict the probability of intensive care unit (ICU) admission and in-hospital death for trauma patients once they arrive at a hospital. A generalized Bayesian rule is used to mine evidence from historical data. Evidence is profiled using a format of belief distribution, where the belief degrees of different trauma outcomes are assigned with derived probabilities linked to the corresponding outcomes. Inputs to the CDSS are clinical data of a patient, and output from the system is predicted belief degree of severe trauma, including ICU admission and in-hospital death. The inner logic of the CDSS is that pieces of evidence that match the clinical data of a patient are identified from the evidence base first, and then the ER rule-based evidence aggregation mechanism is utilized to combine the matched evidences to arrive at a prediction. The reliability, weight, and interdependence of clinical evidence are taken into account. Moreover, an evidence weight training module is constructed. The ER rule-based prediction model has superior performance compared with logistic regression and artificial neural network models. An innovative and pragmatic ER rule-based CDSS for trauma outcome prediction is contributed by this article. In the era of big data, this CDSS helps predict patient outcomes based on historical data and helps physicians in emergency departments make proper trauma management decisions.

Knowledge Base and Rule Base Safety Management Systems Development Strategies and Applications for Petrochemical Industries: A Brief Outline

This review paper discussed about the development tactics and applications of knowledge based and rule-based decision support systems for health and safety in oil and gas sector. It recommends the users on the extent of health and safety concerns for risk and hazard investigation and prevention in which the system developers and researchers mostly face issues during selection and adaptation of development approaches. A complete and systematic review and classification of articles from last seven years has been conducted and briefly analysed, to investigate various methodologies employed in previously developed rule based (RB) and knowledge based (KB) decision support systems. All major bibliographic databases covering a broad series of health and safety fields were searched in various oil and gas production sector related studies. Eighteen articles met the KB and RB study. Twelve of these met the methodological prospective of required investigation. In this review paper, researchers briefly explain about the methodological aspects and industrial applications of rule and knowledge-based decision support systems for reducing work place risks and hazards in safety and health domain. Conclusions that can be extracted from this analytical review paper of the methodological variation and applications of rule based and knowledge-based decision support system is that both inferences are interrelated to each other. RB and KB inferences techniques and methods are used in both interpretations for developing decision support systems in the context of oil and gas extraction and production sector. But safety and health consider KB decision systems as more effective for taking suitable decisions due to real time information and experts’ opinions which facilities safety professionals to take appropriate decision then RB reasoning decision support systems.

A Fuzzy Skill Predictor for Early Childhood Educators

This study presents a model of a two-input single output (TISO) Fuzzy Skill Predictor based on Howard Gardner’s theory of multiple intelligences to assist early childhood educators in discovering latent skills in children of early school age as to tailor them towards professional skill development in their future lives. The skill prediction system was developed in two phases beginning with the generation of weighted fuzzy rules and then followed by the development of a fuzzy rule-based decision support system. The Mamdani Fuzzy inference model in MATLAB was used in implementing the system using weighted attributes of intelligence and ability to determine skills. The system was tested with hypothetical data based on Howard Gardner’s theory of multiple intelligence and was found useful for predicting skills based on the parameters used. The system was validated using early school academic records of 7 randomly sampled undergraduates studying various courses in the university. Though limited entries were used to test the system, the model is robust and can be easily modified to accommodate more entries and rules to predict as many skills as possible.

A consistency analysis-based rule activation method for extended belief-rule-based systems

Problems with inconsistency and incompleteness are widely found in rule-based decision support systems. These problems often impact the accuracy and usability of rule-based decision support systems. The present work focuses on an advanced rule-based decision support system, namely the extended belief-rule-based (EBRB) system, and proposes the consistency analysis-based rule activation (CABRA) method to overcome the above two problems simultaneously. However, two challenges must be discussed and addressed for the EBRB system. First, rather than using activated weights, suitable activated rules must be redefined to better analyze the problems of inconsistency and incompleteness. Second, suitable activated rules must be selected without having to depend on subjective information. Therefore, the proposed CABRA method uses the set of consistent rules as an activation framework to define suitable activated rules before calculating their activated weights, and utilizes the CCR model as a selection model to select suitable activated rules from the set of consistent rules. As such, by embedding the CABRA method, the EBRB system can overcome the problems of inconsistency and incompleteness. Three case studies demonstrate how the use of the CABRA method improves the accuracy and rule activation rate of the EBRB system, which is further confirmed by comparisons with the results of other existing studies. In addition to the work performed in the EBRB system, the CABRA method is treated as a generic rule activation method that can be available for other rule-based decision support systems.

An Efficient Rule based Decision Support System using Semantic Web Technology

The Semantic Web technology is an efficient mechanism to query or infer knowledge on a global scale using the internet by providing logical reasoning through rule based system. In this paper application of semantic web technology is discussed in context of agriculture knowledge management and delivery. In agriculture, adoption of newly developed technology is essential to enhance crop production. However, timely dissemination of authenticated agriculture information for decision making at larger scale to diversified end user has always been a challenge due to several reasons. One of the reasons is storing and delivering agriculture knowledge in machine readable form. In this paper a frame work based on semantic web is presented for collection, storing and updating of agricultural information at centralized location and delivering knowledge through intelligent decision support system through semantic web. The frame work utilizes rule based system for querying information from agriculture knowledge base.

The Application of Aeronautical Decision-making Support Systems for Improving Pilots' Performance in Flight Operations

Operating a high-technology commercial airliner is not only an issue in psychomotor skill performance but also of a real-time decision-making involving situation awareness and risk management within a limited-time condition. The number of aircraft accidents attributable solely to mechanical failures has decreased markedly in recent years, but the contribution of human error has declined at a much slower rate. Previous research demonstrated a belief rule-based decision support system has provided more reliable and informative performance after training. The purpose of this research was to identify the best mnemonic-based method of decision support systems for improving commercial pilot's performance in the advanced cockpit. A total of 157 airline pilots, all qualified on the Boeing 747-400 evaluated the suitability of four different ADM methods: SHOR (Stimuli, Hypotheses, Options, Response); PASS (Problem identification, Acquire information, Survey strategy, Select strategy); FORDEC (Facts, Options, Risks & Benefits, Decision, Execution, Check); and DESIDE (Detect, Estimate, Set safety objectives, Identify, Do, Evaluate). Each was evaluated for six different types of decisions: go/no go; recognition-primed; response selection; resource management; non-diagnostic procedural; and creative problem-solving. Pilots regarded the FORDEC methodology as being the best in all decision-making scenarios, irrespective of the time available to make the decision. It was also rated as the best ADM method for promoting crew coordination. However, it was advised that practicing the FORDEC mnemonic in flight simulator was important before attempting to apply it in a real life situation.

Abstract 018: Prior Authorization for Diagnostic Catheterization: The Value of Reviewers When There is Clinical Ambiguity

Background: There are many clinical situations in which evidence-based guidelines cannot definitively determine the appropriateness of diagnostic catheterization. One specialty benefits management company has taken a two-step approach towards handling this ambiguity: first evaluating the appropriateness of orders using a rule-based decision support system, and then having reviewers provide additional input through the review process of a nondenial prior authorization program. When reviewers did not find a clear case for appropriateness, ordering physicians were engaged in a discussion with a peer physician. This program evaluation reports on the outcomes of diagnostic catheterization orders analyzed through this two-step process. Methods: Data from 2015 elective diagnostic catheterization orders submitted by one health insurer’s Medicare Advantage plans were used for this program evaluation. The rates at which the rule-based system suggested orders were inadequately justified, potentially nonindicated, and potentially appropriate are presented. Rates of approval after review and information gathering by the review process are presented for the three groups of orders. Chi-squared tests were conducted to examine whether the classification of the orders by the rule-based system and the review system were independent of the plan type (HMO or PPO), the specialty of the ordering physician (cardiologist or noncardiologist), or state of residence of the patient (FL, KY, LA, OH, TX, or other). Results: There were 3,808 orders for elective diagnostic catheterization, and inadequate initial justification was provided for 699 (18.4%) of the orders. After inquiry through the review process, 509 (72.8%) of these orders were approved. Among the 344 (9.0%) orders deemed potentially nonindicated by the rule-based system, the review process approved 298 (86.7%). Of the 2,765 (72.6%) orders deemed potentially appropriate by the rule-based system, the review process approved 99.1% (2,740/2,765). Chi-squared tests did not show a significant association between plan type (p=0.18) or physician specialty (p=0.89) and the classification of the order by the rule-based system. However, there was a significant association between the classification of the order by the rule-based system and state of residence of the patient (p<0.001). There was not a significant association between the outcome of the review process and the health plan type (p=0.10), the provider’s specialty (p=0.57), or the state of residence of the patient (p=0.73). Conclusions: Rule-based decision support can be combined with a review process featuring peer discussion to determine whether care is appropriate when guidelines are ambiguous. Orders which are poorly justified are often supportable after gathering information on the patient’s presentation. There may be state-based variation in the appropriateness of orders.

Screening Consolidated Clinical Document Architecture (CCDA) Documents for Sensitive Data Using a Rule-Based Decision Support System.

The Centers for Medicare & Medicaid Services' Stage 2 final rule requires that eligible hospitals provide a visit summary electronically at transitions of care in order to qualify for "meaningful use" incentive payments. However, Massachusetts state law and Federal law prohibit the transmission of documents containing "sensitive" data unless there is a new patient consent for each transmission. To describe the implementation and evaluation of a rule-based decision support system used to screen transition of care documents for sensitive data. We implemented a rule-based document screening system to identify transition of care documents that might contain sensitive data. The transmission of detected documents is withheld until a new patient consent is obtained. The documents that were flagged as containing sensitive data were reviewed in two different time periods to verify that the decision support system was not missing documents or withholding more documents than necessary. The rule-based screening system has been in regular production use for the past 18 months. During the first evaluation period, 3% of 5,841 documents were identified as containing sensitive data (true-positive rate of 44%). After additional enhancements to the rules, the system was evaluated a second time and 4.5% of 6,935 documents were identified as containing sensitive data (true-positive rate of 98.4%). The analysis of the system demonstrates that production rules can be used to automatically screen the content of transition of care documents for sensitive data. The utilization of the rule-based decision support system enabled our hospitals to achieve meaningful use and, at the same time, remain compliant with state and federal laws.