Diagnostic Decision Support System Research Articles

An ontology-driven, diagnostic modeling system.

To present a system that uses knowledge stored in a medical ontology to automate the development of diagnostic decision support systems. To illustrate its function through an example focused on the development of a tool for diagnosing pneumonia. We developed a system that automates the creation of diagnostic decision-support applications. It relies on a medical ontology to direct the acquisition of clinic data from a clinical data warehouse and uses an automated analytic system to apply a sequence of machine learning algorithms that create applications for diagnostic screening. We refer to this system as the ontology-driven diagnostic modeling system (ODMS). We tested this system using samples of patient data collected in Salt Lake City emergency rooms and stored in Intermountain Healthcare's enterprise data warehouse. The system was used in the preliminary development steps of a tool to identify patients with pneumonia in the emergency department. This tool was compared with a manually created diagnostic tool derived from a curated dataset. The manually created tool is currently in clinical use. The automatically created tool had an area under the receiver operating characteristic curve of 0.920 (95% CI 0.916 to 0.924), compared with 0.944 (95% CI 0.942 to 0.947) for the manually created tool. Initial testing of the ODMS demonstrates promising accuracy for the highly automated results and illustrates the route to model improvement. The use of medical knowledge, embedded in ontologies, to direct the initial development of diagnostic computing systems appears feasible.

An extensible six-step methodology to automatically generate fuzzy DSSs for diagnostic applications

BackgroundThe diagnosis of many diseases can be often formulated as a decision problem; uncertainty affects these problems so that many computerized Diagnostic Decision Support Systems (in the following, DDSSs) have been developed to aid the physician in interpreting clinical data and thus to improve the quality of the whole process. Fuzzy logic, a well established attempt at the formalization and mechanization of human capabilities in reasoning and deciding with noisy information, can be profitably used. Recently, we informally proposed a general methodology to automatically build DDSSs on the top of fuzzy knowledge extracted from data.MethodsWe carefully refine and formalize our methodology that includes six stages, where the first three stages work with crisp rules, whereas the last three ones are employed on fuzzy models. Its strength relies on its generality and modularity since it supports the integration of alternative techniques in each of its stages.ResultsThe methodology is designed and implemented in the form of a modular and portable software architecture according to a component-based approach. The architecture is deeply described and a summary inspection of the main components in terms of UML diagrams is outlined as well. A first implementation of the architecture has been then realized in Java following the object-oriented paradigm and used to instantiate a DDSS example aimed at accurately diagnosing breast masses as a proof of concept.ConclusionsThe results prove the feasibility of the whole methodology implemented in terms of the architecture proposed.

Selecting Relevant Information for Medical Decision Support with Application in Cardiology

Objectives: The aim of our work was to implement a prototype of a decision support system which has the form of a web-based classification service. Because the data analysis component of decision support systems often happens to be unsuitable for high-dimensional data, special attention must be paid to the sophisticated selection of the most relevant variables before learning the classification rule. Methods: We implemented a prototype of a diagnostic decision support system called SIR. The system has the ability to select the most relevant variables based on a set of high-dimensional measurements by means of a forward procedure optimizing a decision-making criterion. This allows to learn a reliable classification rule. Results: The implemented prototype was tested on a sample of patients involved in a cardiology study. We used SIR to perform an information extraction from a cardiological clinical study containing both clinical and gene expression data. The classification performance was evaluated by means of a cross validation study. Conclusions: The proposed classification system can be useful for clinicians in primary care to support their decision-making tasks with relevant information extracted from any available clinical study. It is especially suitable for analyzing high-dimensional data, e.g. gene expression measurements.

Towards an Ontology to Support Semantics Enabled Diagnostic Decision Support Systems

Towards an Ontology to Support Semantics Enabled Diagnostic Decision Support Systems

Impact of a Computer-Based Diagnostic Decision Support Tool on the Differential Diagnoses of Medicine Residents

Computer-based medical diagnostic decision support systems have been used for decades, initially as stand-alone applications. More recent versions have been tested for their effectiveness in enhancing the diagnostic ability of clinicians. To determine if viewing a rank-ordered list of diagnostic possibilities from a medical diagnostic decision support system improves residents' differential diagnoses or management plans. Twenty first-year internal medicine residents at Massachusetts General Hospital viewed 3 deidentified case descriptions of real patients. All residents completed a web-based questionnaire, entering the differential diagnosis and management plan before and after seeing the diagnostic decision support system's suggested list of diseases. In all 3 exercises, the actual case diagnosis was first on the system's list. Each resident served as his or her own control (pretest/posttest). For all 3 cases, a substantial percentage of residents changed their primary considered diagnosis after reviewing the system's suggested diagnoses, and a number of residents who had not initially listed a "further action" (laboratory test, imaging study, or referral) added or changed their management options after using the system. Many residents (20% to 65% depending on the case) improved their differential diagnosis from before to after viewing the system's suggestions. The average time to complete all 3 cases was 15.4minutes. Most residents thought that viewing the medical diagnostic decision support system's list of suggestions was helpful. Viewing a rank-ordered list of diagnostic possibilities from a diagnostic decision support tool had a significant beneficial effect on the quality of first-year medicine residents' differential diagnoses and management plans.

A Knowledge-based Diagnostic Clinical Decision Support System for Musculoskeletal Disorders of the Shoulder for Use in a Primary Care Setting

Background Twenty percent of cases seen by primary care clinicians (general practitioners; GPs) are musculoskeletal in nature, and approximately one-quarter of these are shoulder complaints. GPs are increasingly overloaded with clinical information and unfamiliarity with current research can easily lead to misdiagnosis and, in turn, to unnecessary test requests or onward specialist referrals. Well-designed diagnostic clinical decision support systems (CDSS) have been shown to facilitate clinical decision-making and reduce diagnostic errors. However, no CDSS have been developed or tested for musculoskeletal disorders. Methods We have developed a prototype knowledge-based diagnostic CDSS for musculoskeletal shoulder conditions. The CDSS uses Bayesian reasoning to diagnose six common musculoskeletal shoulder pathologies, based on current evidence and expert opinion. The CDSS was tested by comparing its diagnostic outcome against 50 case studies with known diagnosis by radiological imaging. Results The CDSS diagnostic validity and reliability was shown to be 88% with a Kappa value of 0.85 to a confidence level of 99% compared to known diagnosis by radiological imaging. Conclusions The results suggest that a Bayesian network-based CDSS is a promising instrument in the diagnosis of musculoskeletal shoulder conditions, having been shown to be valid and reliable for 50 case studies.

Presence of key findings in the medical record prior to a documented high-risk diagnosis.

Failure or delay in diagnosis is a common preventable source of error. The authors sought to determine the frequency with which high-information clinical findings (HIFs) suggestive of a high-risk diagnosis (HRD) appear in the medical record before HRD documentation. A knowledge base from a diagnostic decision support system was used to identify HIFs for selected HRDs: lumbar disc disease, myocardial infarction, appendicitis, and colon, breast, lung, ovarian and bladder carcinomas. Two physicians reviewed at least 20 patient records retrieved from a research patient data registry for each of these eight HRDs and for age- and gender-compatible controls. Records were searched for HIFs in visit notes that were created before the HRD was established in the electronic record and in general medical visit notes for controls. 25% of records reviewed (61/243) contained HIFs in notes before the HRD was established. The mean duration between HIFs first occurring in the record and time of diagnosis ranged from 19 days for breast cancer to 2 years for bladder cancer. In three of the eight HRDs, HIFs were much less likely in control patients without the HRD. In many records of patients with an HRD, HIFs were present before the HRD was established. Reasons for delay include non-compliance with recommended follow-up, unusual presentation of a disease, and system errors (eg, lack of laboratory follow-up). The presence of HIFs in clinical records suggests a potential role for the integration of diagnostic decision support into the clinical workflow to provide reminder alerts to improve the diagnostic focus.

Analysis of a Multilevel Diagnosis Decision Support System and Its Implications: A Case Study

Medical diagnosis can be performed in an automatic way with the use of computer-based systems or algorithms. Such systems are usually called diagnostic decision support systems (DDSSs) or medical diagnosis systems (MDSs). An evaluation of the performance of a DDSS called ML-DDSS has been performed in this paper. The methodology is based on clinical case resolution performed by physicians which is then used to evaluate the behavior of ML-DDSS. This methodology allows the calculation of values for several well-known metrics such as precision, recall, accuracy, specificity, and Matthews correlation coefficient (MCC). Analysis of the behavior of ML-DDSS reveals interesting results about the behavior of the system and of the physicians who took part in the evaluation process. Global results show how the ML-DDSS system would have significant utility if used in medical practice. The MCC metric reveals an improvement of about 30% in comparison with the experts, and with respect to sensitivity the system returns better results than the experts.

E-learning - evaluation of a diagnostic decision support system in undergraduate medical training

Abstract Background Enhanced diagnostic reasoning by medical students has a positive effect on learning diagnostic procedures. Decision support systems (DSS) are intended to support this process in the clinical phase of medical education. The aim was to explore both the use and usefulness of such a system, as well as adjustments needed for an optimal implementation in undergraduate medical training. Methods Using a dispersive implementation strategy, fourth-year medical students (n=130) were given access to a DSS (DXplain) during the two-semester course in internal medicine. The students were asked to participate in an internet-based questionnaire survey at the end. Results A majority (61.8%) of the users found the system more or less useful, 41.8% found it relatively easy to use, and a majority (54.9%) would like to have future access to the system. However, the usage was limited, and several issues needed for an optimal implementation were found related to interface, logic, usefulness, accessibility and motivation. Conclusions The DXplain system were more or less useful and relatively easy to use in undergraduate medical training, and the usage was limited using an dispersive implementation. An optimal implementation needs to be stricter including taking several aspects of usefulness into account, and a pronounced integration in theoretical and clinical phases of the education is needed to take advantage of its full potential in undergraduate medical training.

Cross-domain probabilistic inference in a clinical decision support system: Examples for dermatology and rheumatology

Introduction Maintaining a large diagnostic knowledge base (KB) is a demanding task for any person or organization. Future clinical decision support system (CDSS) may rely on multiple, smaller and more focused KBs developed and maintained at different locations that work together seamlessly. A cross-domain inference tool has great clinical import and utility. Methods We developed a modified multi-membership Bayes formulation to facilitate the cross-domain probabilistic inferencing among KBs with overlapping diseases. Two KBs developed for evaluation were non-infectious generalized blistering diseases (GBD) and autoimmune diseases (AID). After the KBs were finalized, they were evaluated separately for validity. Result Ten cases from medical journal case reports were used to evaluate this “cross-domain” inference across the two KBs. The resultant non-error rate (NER) was 90%, and the average of probabilities assigned to the correct diagnosis (AVP) was 64.8% for cross-domain consultations. Conclusion A novel formulation is now available to deal with problems occurring in a clinical diagnostic decision support system with multi-domain KBs. The utilization of this formulation will help in the development of more integrated KBs with greater focused knowledge domains.

Modern parameterization and explanation techniques in diagnostic decision support system: A case study in diagnostics of coronary artery disease

Coronary artery disease has been described as one of the curses of the western world, as it is one of its most important causes of mortality. Therefore, clinicians seek to improve diagnostic procedures, especially those that allow them to reach reliable early diagnoses. In the clinical setting, coronary artery disease diagnostics are typically performed in a sequential manner. The four diagnostic levels consist of evaluation of (1) signs and symptoms of the disease and electrocardiogram at rest, (2) sequential electrocardiogram testing during the controlled exercise, (3) myocardial perfusion scintigraphy, and (4) finally coronary angiography, that is considered as the "gold standard" reference method. Our study focuses on improving diagnostic performance of the third, virtually non-invasive, diagnostic level. Myocardial scintigraphy results in a series of medical images that are obtained by relatively inexpensive means. In clinical practice, these images are manually described (parameterized) by expert physicians. In the paper we present an innovative alternative to manual image evaluation-an automatic image parameterization on multiple resolutions, based on texture description with specialized association rules. Extracted image parameters are combined into more informative composite parameters by means of principal component analysis, and finally used to build automatic classifiers with machine learning methods. Our experiments with synthetic datasets show that association-rule-based multi-resolution image parameterization works very well for scintigraphic images of the heart. In coronary artery disease diagnostics we confirm these results as our approach significantly improves on clinical results in terms of diagnostic performance. We improve diagnostic accuracy by 17%, specificity by 12% and sensitivity by 22%. We also significantly improve the number of reliably diagnosed patients by 19% for positive diagnoses, and 16% for negative diagnoses, so that no costly further tests are necessary for them. Multi-resolution image parameterization equals or even betters that of the physicians in terms of the diagnostic quality of image parameters. By using these parameters for building machine learning classifiers, we can significantly improve diagnostic performance with respect to the results of clinical practice, affect process rationalization, as well as possibly provide novel insights into the diagnostic problems, features and/or processes.

Diagnostic accuracy in patients admitted to hospitals with cellulitis

Misdiagnosis of non-infectious conditions such as cellulitis is a common error and can result in unnecessary hospitalization and antibiotic use. We sought to prospectively determine the misdiagnosis rate of cellulitis among hospitalized patients and to determine if a visually-based computerized diagnostic decision support system (VCDDSS, also named VisualDx) could generate an improved differential diagnosis (DDx) for misdiagnosed patients. In two separate institutions, attending dermatologists or infectious disease specialists evaluated all consecutive patients hospitalized for "cellulitis" by the emergency department. Among 145 subjects enrolled, misdiagnosis occurred in 41 (28%) patients. The diagnosis most commonly mistaken as cellulitis was stasis dermatitis (37%). At one center, in cases that were misdiagnosed by the emergency department, the VCDDSS included the correct diagnosis in the DDx more frequently than the admitting team (18/28 cases (64%) compared to 4/28 cases (14%), p=0.0003). These results demonstrate the capability of this VCDDSS to assist primary care physicians with generating a more accurate DDx when confronted with patients presenting with possible skin infections. Misdiagnoses may result in a significant source of healthcare costs and misdiagnosis-related patient harm. Inclusion of decision support tools early in the diagnostic workflow may reduce misdiagnosis and result in more efficient healthcare management.

Experts' knowledge acquisition for differential diagnostics of bronchial asthma in children in STEPCLASS environment

The problem of experts' knowledge acquisition for diagnostic decision support systems (DDSS) is considered as a multi-attribute classification problem (MACP), where a class is the subset of objects with the same diagnosis. Method STEPCLASS has proved to be an effective tool for such problem solving in various applications due to its ability to construct complete (up to the experts' knowledge) and contradiction-free knowledge bases. The main techniques of STEPCLASS – application domain structuring, large-size problem decomposition, classification rules eliciting, and rules' inconsistency control – are considered and illustrated with its application for constructing knowledge base of DDSS for differential diagnostics of bronchial asthma in children (DDBAC).

Clasification Of Arrhythmic ECG Data Using Machine Learning Techniques

In this paper we proposed a automated Artificial Neural Network (ANN) based classification system for cardiac arrhythmia using multi-channel ECG recordings. In this study, we are mainly interested in producing high confident arrhythmia classification results to be applicable in diagnostic decision support systems. Neural network model with back propagation algorithm is used to classify arrhythmia cases into normal and abnormal classes. Networks models are trained and tested for MIT-BIH arrhythmia. The different structures of ANN have been trained by mixture of arrhythmic and non arrhythmic data patient. The classification performance is evaluated using measures; sensitivity, specificity, classification accuracy, mean squared error (MSE), receiver operating characteristics (ROC) and area under curve (AUC). Our experimental results gives 96.77% accuracy on MIT-BIH database and 96.21% on database prepared by including NSR database also.

The introduction of a diagnostic decision support system (DXplain™) into the workflow of a teaching hospital service can decrease the cost of service for diagnostically challenging Diagnostic Related Groups (DRGs)

Background In an era of short inpatient stays, residents may overlook relevant elements of the differential diagnosis as they try to evaluate and treat patients. However, if a resident's first principal diagnosis is wrong, the patient's appropriate evaluation and treatment may take longer, cost more, and lead to worse outcomes. A diagnostic decision support system may lead to the generation of a broader differential diagnosis that more often includes the correct diagnosis, permitting a shorter, more effective, and less costly hospital stay. Methods We provided residents on General Medicine services access to DXplain, an established computer-based diagnostic decision support system, for 6 months. We compared charges and cost of service for diagnostically challenging cases seen during the fourth through sixth month of access to DXplain (intervention period) to control cases seen in the 6 months before the system was made available. Results 564 cases were identified as diagnostically challenging by our criteria during the intervention period along with 1173 cases during the control period. Total charges were $1281 lower ( p = .006), Medicare Part A charges $1032 lower ( p = 0.006) and cost of service $990 lower ( p = 0.001) per admission in the intervention cases than in control cases. Conclusions Using DXplain on all diagnostically challenging cases might save our medical center over $2,000,000 a year on the General Medicine Services alone. Using clinical diagnostic decision support systems may improve quality and decrease cost substantially at teaching hospitals.

Automated decision algorithms in prostate cancer

A diagnostic decision support system (DDSS) is defi ned as a methodology that provides guidance in situations involving complex decision sequences. DDSSs result in a systematic, ordered, and exhaustive evaluation of evidence and weighting of individual items of evidence as they are combined to form the basis for a fi nal decision. Most DDSSs provide a numeric measure of confi dence in the fi nal decision or diagnostic recommendation. Th e decisive advantage of DDSSs is the ability to process descriptive symbolic information, in contrast to systems limited to the handling of numerical information only for which extensive analytical procedures are already established. Most human knowledge and insight related to diagnostic and prognostic evaluation exist in symbolic form as concepts and linguistic terms, so the DDSSs have facilitated systematic evaluation of evidence to provide diagnostic and prognostic decision support. DDSSs may be implemented as inference networks (or Bayesian Belief Networks – BBNs), automated reasoning systems, case-based reasoning systems, or expert systems. In inference networks and automated reasoning systems, the emphasis is on uncertainty assessment of a given decision sequence. In case-base reasoning, the emphasis is on prognostic assessment for an individual patient. In expert systems, the emphasis is on diagnostic or prognostic assessment, by making available a comprehensive knowledge base of facts and professional experience. However, even though the emphasis is slightly diff erent in these kinds of decision support systems, much of the methodology is shared. A BBN consists of a decision node for the diagnostic alternatives and of evidence nodes for the diagnostic clues. Each clue is observed, rated, and assigned to a function with a given probability. Th e evidence is then forwarded to the decision node via a conditional probability link matrix. At the decision node, the belief in each diagnostic alternative is accumulated. A series of BBNs have already been successfully developed for prostate and non-prostate neoplasms (7-13) including comparing cancer with benign lesions and those with prostatic intraepithelial neoplasms.

Domain structuring for knowledge-based multiattribute classification (a verbal decision analysis approach)

Classification problems in decision making are, at least, ill-structured or even unstructured ones, since, among other things, human judgments (i.e., Decision Maker preferences and/or expert knowledge) are the primary sources of information for their solving. Thus, not only the classification rules eliciting, but the application domain structuring as well, is a complex problem itself. The paper focuses on knowledge-based classification problem structuring in the context of complete (up to the expert knowledge) and consistent knowledge base construction for a Diagnostic Decision Support System. Two structuring techniques are proposed as expert aids, as well as an approach to large-size problem decomposition. It is asserted that application domain structuring and classification rules eliciting have to be arranged as interconnected procedures.

Use of Diagnostic Decision Support Systems in Medical Education

Diagnostic decision support systems are designed to assist physicians with making diagnoses. This article illustrates some of the issues that will be faced as diagnostic decision support systems become used in medical education. The objectives of this article are to examine 1) the skills that are needed to properly use these programs as part of the students' clinical experiences; 2) the changes that will be necessary in our curricula once these programs are more extensively utilized, including the implications of using these systems as an educational resource or simulation tool, and 3) the research issues that arise when these systems become an established part of our educational programs. This is a critical analysis of the literature on diagnostic decision support systems and medical education. To optimally use diagnostic decision support programs, students will need grounding in the basic knowledge and skills that have always been necessary to become a physician, such as the ability to accurately gather and interpret clinical information from the patient. In addition, students will need specific skills in 1) selecting appropriate system vocabulary and functions, and 2) applying the diagnostic system's suggestions to their particular patient. When computer-based decision support systems are incorporated in medical education, they will likely lead to changes in the traditional medical curriculum. Research will be needed on how use of these programs changes the students' knowledge, problem-solving and information-seeking skills.

61st Annual Meeting of the American Academy of Neurology Seattle, Washington April 25-May 2, 2009

61st Annual Meeting of the American Academy of Neurology Seattle, Washington April 25-May 2, 2009

Computer-assisted diagnostic decision support: history, challenges, and possible paths forward

This paper presents a brief history of computer-assisted diagnosis, including challenges and future directions. Some ideas presented in this article on computer-assisted diagnostic decision support systems (CDDSS) derive from prior work by the author and his colleagues (see list in Acknowledgments) on the INTERNIST-1 and QMR projects. References indicate the original sources of many of these ideas.