Search

At the University of Miyazaki Hospital (UMH), we have accumulated and semantically structured a vast amount of medical information since the activation of the electronic health record system approximately 10 years ago. With this medical information, we have decided to develop an alert system for aiding in medical treatment. The purpose of this investigation is to not only to integrate an alert framework into the electronic heath record system, but also to formulate a modeling method of this knowledge. A trial alert framework was developed for the staff in various occupational categories at the UMH. Based on findings of subsequent interviews, a more detailed and upgraded alert framework was constructed, resulting in the final model. Based on our current findings, an alert framework was developed with four major items. Based on the analysis of the medical practices from the trial model, it has been concluded that there are four major risk patterns that trigger the alert. Furthermore, the current alert framework contains detailed definitions which are easily substituted into the database, leading to easy implementation of the electronic health records.

The primary diagnosis of Tuberculosis (TB) is usually carried out by looking at the various signs and symptoms of a patient. However, these signs and symptoms cannot be measured with 100 % certainty since they are associated with various types of uncertainties such as vagueness, imprecision, randomness, ignorance and incompleteness. Consequently, traditional primary diagnosis, based on these signs and symptoms, which is carried out by the physicians, cannot deliver reliable results. Therefore, this article presents the design, development and applications of a Belief Rule Based Expert System (BRBES) with the ability to handle various types of uncertainties to diagnose TB. The knowledge base of this system is constructed by taking experts’ suggestions and by analyzing historical data of TB patients. The experiments, carried out, by taking the data of 100 patients demonstrate that the BRBES’s generated results are more reliable than that of human expert as well as fuzzy rule based expert system.

In the present era, soft computing approaches play a vital role in solving the different kinds of problems and provide promising solutions. Due to popularity of soft computing approaches, these approaches have also been applied in healthcare data for effectively diagnosing the diseases and obtaining better results in comparison to traditional approaches. Soft computing approaches have the ability to adapt itself according to problem domain. Another aspect is a good balance between exploration and exploitation processes. These aspects make soft computing approaches more powerful, reliable and efficient. The above mentioned characteristics make the soft computing approaches more suitable and competent for health care data. The first objective of this review paper is to identify the various soft computing approaches which are used for diagnosing and predicting the diseases. Second objective is to identify various diseases for which these approaches are applied. Third objective is to categories the soft computing approaches for clinical support system. In literature, it is found that large number of soft computing approaches have been applied for effectively diagnosing and predicting the diseases from healthcare data. Some of these are particle swarm optimization, genetic algorithm, artificial neural network, support vector machine etc. A detailed discussion on these approaches are presented in literature section. This work summarizes various soft computing approaches used in healthcare domain in last one decade. These approaches are categorized in five different categories based on the methodology, these are classification model based system, expert system, fuzzy and neuro fuzzy system, rule based system and case based system. Lot of techniques are discussed in above mentioned categories and all discussed techniques are summarized in the form of tables also. This work also focuses on accuracy rate of soft computing technique and tabular information is provided for each category including author details, technique, disease and utility/accuracy.

In healthcare organizations, clinical workflows are executed by interdisciplinary healthcare teams (IHTs) that operate in ways that are difficult to manage. Responding to a need to support such teams, we designed and developed the MET4 multi-agent system that allows IHTs to manage patients according to presentation-specific clinical workflows. In this paper, we describe a significant extension of the MET4 system that allows for supporting rich team dynamics (understood as team formation, management and task-practitioner allocation), including selection and maintenance of the most responsible physician and more complex rules of selecting practitioners for the workflow tasks. In order to develop this extension, we introduced three semantic components: (1) a revised ontology describing concepts and relations pertinent to IHTs, workflows, and managed patients, (2) a set of behavioral rules describing the team dynamics, and (3) an instance base that stores facts corresponding to instances of concepts from the ontology and to relations between these instances. The semantic components are represented in first-order logic and they can be automatically processed using theorem proving and model finding techniques. We employ these techniques to find models that correspond to specific decisions controlling the dynamics of IHT. In the paper, we present the design of extended MET4 with a special focus on the new semantic components. We then describe its proof-of-concept implementation using the WADE multi-agent platform and the Z3 solver (theorem prover/model finder). We illustrate the main ideas discussed in the paper with a clinical scenario of an IHT managing a patient with chronic kidney disease.

The use of expert systems and artificial intelligence techniques in disease diagnosis has been increasing gradually. Artificial Immune Recognition System (AIRS) is one of the methods used in medical classification problems. AIRS2 is a more efficient version of the AIRS algorithm. In this paper, we used a modified AIRS2 called MAIRS2 where we replace the K- nearest neighbors algorithm with the fuzzy K-nearest neighbors to improve the diagnostic accuracy of diabetes diseases. The diabetes disease dataset used in our work is retrieved from UCI machine learning repository. The performances of the AIRS2 and MAIRS2 are evaluated regarding classification accuracy, sensitivity and specificity values. The highest classification accuracy obtained when applying the AIRS2 and MAIRS2 using 10-fold cross-validation was, respectively 82.69% and 89.10%.

In this study, it is aimed, using the Web-based Expert System with Fuzzy Input (WESFI), to convert the patients' (users') Serology Laboratory Tests (SLT) results to linguistic statements (low, normal, high) and analyzing those, give a feedback to the user (patient) of the potential signs of disease. The feedbacks given to the patients are the existing interpretations in the database, which were prepared by doctors before. Furthermore, the SLT terms (Brucella Coombs, Ama, P-Protein etc.) are explained in a way that the user can understand. The WESFI is published with an interface on the web environment. In order to determine the rate of the success of the WESFI, users evaluated the system answering the "How do you find the evaluation?" question. The question has been answered by 461 users. As a result it is observed that 90% of female users, 92% of male users and 91% of all users found the system useful.

In nowadays, there are many various diseases, whose diagnosis is very hardly. Lung cancer is one of this type diseases. It begins in the lungs and spreads to other organs of human body. In this paper, an expert diagnostic system based on General Discriminant Analysis (GDA) and Least Square Support Vector Machine (LS-SVM) Classifier for diagnosis of lung cancer. This expert diagnosis system is called as GDA-LS-SVM in rest of this paper. The GDA-LS-SVM expert diagnosis system has two stages. These are 1. Feature extraction and feature reduction stage and 2. Classification stage. In feature extraction and feature reduction stage, lung cancer dataset is obtained and dimension of this lung cancer dataset, which has 57 features, is reduced to eight features using Generalized Discriminant Analysis (GDA) method. Then, in classification stage, these reduced features are given to Least Squares Support Vector Machine (LS-SVM) classifier. The lung cancer dataset used in this study was taken from the UCI machine learning database. The classification accuracy of this GDA-LS-SVM expert system was obtained about 96.875% from results of these experimental studies.

This paper describes the development of a Chinese Acupuncture Expert System (CAES) that will assist the learning and practice of medical acupuncture. This was the development of a Chinese Acupuncture Expert System which incorporated eight functional modules. These modules were 1) Add Patient Record subsystem; 2) Diagnosis subsystem ; 3) Acupuncture Prescription subsystem ; 4) Needle Insertion Position Animation subsystem ; 5) Acupuncture Points Usage Statistic subsystem ; 6) History Query subsystem; 7) Acupuncture Points Query subsystem and 8) Diagnosis Remarks and Diagnosis Record Save subsystem. Two databases were built--Patient Record database and Diagnosis (Acupuncture) Knowledge database. All the Traditional Chinese Medicine (TCM) knowledge and acupuncture treatment prescriptions were extracted from officially used TCM textbooks and received guidiance and expert advice from two acupuncturists working in this project. A Chinese Acupuncture Expert System (CAES) was built, which after the input from users of any Chinese disease symptoms and signs, it can provide a list of related TCM syndrome diagnoses based on the patients' disease symptoms and signs, and at the same time it can offer advice of the appropriate Chinese acupuncture treatment to the users. CAES also provided text descriptions and acupuncture animations showing the acupoint locations and the direction and depth of the needle insertion technique. Therefore users can easily learn acupuncture and practice it. This new acupuncture expert system will hopefully provide an easy way for users to learn and practice Chinese Acupuncture and establish its usefulness after it was fully evaluated.

Prescription medicine for asthma at primary stages is based on asthma severity level. Despite major progress in discovering various variables affecting asthma severity levels, disregarding some of these variables by physicians, variables' inherent uncertainty, and assigning patients to limited categories of decision making are the major causes of underestimating asthma severity, and as a result low quality of life in asthmatic patients. In this paper, we provide a solution of intelligence fuzzy system for this problem. Inputs of this system are organized in five modules of respiratory symptoms, bronchial obstruction, asthma instability, quality of life, and asthma severity. Output of this system is degree of asthma severity in score (0---10). Evaluating performance of this system by 28 asthmatic patients reinforces that the system's results not only correspond with evaluations of physicians, but represent the slight differences of asthmatic patients placed in specific category introduced by guidelines.

Recent advances in digital imaging technology have greatly enhanced the interpretation of critical/pathology conditions from the 2-dimensional medical images. This has become realistic due to the existence of the computer aided diagnostic tool. A computer aided diagnostic (CAD) tool generally possesses components like preprocessing, identification/selection of region of interest, extraction of typical features and finally an efficient classification system. This paper enumerates on development of CAD tool for classification of chronic liver disease through the 2-D image acquired from ultrasonic device. Characterization of tissue through qualitative treatment leads to the detection of abnormality which is not viable through qualitative visual inspection by the radiologist. Common liver diseases are the indicators of changes in tissue elasticity. One can show the detection of normal, fatty or malignant condition based on the application of CAD tool thereby, further investigation required by radiologist can be avoided. The proposed work involves an optimal block analysis (64 x 64) of the liver image of actual size 256 x 256 by incorporating Gabor wavelet transform which does the texture classification through automated mode. Statistical features such as gray level mean as well as variance values are estimated after this preprocessing mode. A non-linear back propagation neural network (BPNN) is applied for classifying the normal (vs) fatty and normal (vs) malignant liver which yields a classification accuracy of 96.8%. Further multi classification is also performed and a classification accuracy of 94% is obtained. It can be concluded that the proposed CAD can be used as an expert system to aid the automated diagnosis of liver diseases.

The objective of study is to examine factors affecting the variation in technical and cost efficiency of community health centers (CHCs). A panel study design was formulated to examine the relationships among the contextual, organizational structural, and performance variables. Data Envelopment Analysis (DEA) of technical efficiency and latent growth curve modeling of multi-wave technical and cost efficiency were performed. Regardless of the efficiency measures, CHC efficiency was influenced more by contextual factors than organizational structural factors. The study confirms the independent and additive influences of contextual and organizational predictors on efficiency. The change in CHC technical efficiency positively affects the change in CHC cost efficiency. The practical implication of this finding is that healthcare managers can simultaneously optimize both technical and cost efficiency through appropriate use of inputs to generate optimal outputs. An innovative solution is to employ decision support software to prepare an expert system to assist poorly performing CHCs to achieve better cost efficiency through optimizing technical efficiency.

The focus of this paper is diagnosing and differentiating Astrocytomas in MRI scans by developing an interval Type-2 fuzzy automated tumor detection system. This system consists of three modules: working memory, knowledge base, and inference engine. An image processing method with three steps of preprocessing, segmentation and feature extraction, and approximate reasoning is used in inference engine module to enhance the quality of MRI scans, segment them into desired regions, extract the required features, and finally diagnose and differentiate Astrocytomas. However, brain tumors have different characteristics in different planes, so considering one plane of patient's MRI scan may cause inaccurate results. Therefore, in the developed system, several consecutive planes are processed. The performance of this system is evaluated using 95 MRI scans and the results show good improvement in diagnosing and differentiating Astrocytomas.

Complicacy of clinical decisions justifies utilization of information systems such as artificial intelligence (e.g. expert systems and neural networks) to achieve better decisions, however, application of these systems in the medical domain faces some challenges. We aimed at to review the applications of these systems in the medical domain and discuss about such challenges. Following a brief introduction of expert systems and neural networks by representing few examples, the challenges of these systems in the medical domain are discussed. We found that the applications of expert systems and artificial neural networks have been increased in the medical domain. These systems have shown many advantages such as utilization of experts' knowledge, gaining rare knowledge, more time for assessment of the decision, more consistent decisions, and shorter decision-making process. In spite of all these advantages, there are challenges ahead of developing and using such systems including maintenance, required experts, inputting patients' data into the system, problems for knowledge acquisition, problems in modeling medical knowledge, evaluation and validation of system performance, wrong recommendations and responsibility, limited domains of such systems and necessity of integrating such systems into the routine work flows. We concluded that expert systems and neural networks can be successfully used in medicine; however, there are many concerns and questions to be answered through future studies and discussions.

Cardiac events could be taken into account as the leading causes of death throughout the globe. Such events also trigger an undesirable increase in what treatment procedures cost. Despite the giant leaps in technological development in heart surgery, coronary surgery still carries the high risk of the mortality. Besides, there is still a long way ahead to accurately predict and assess the mortality risk. This study is an attempt to develop an expert system for the risk assessment of mortality following the cardiac surgery. The developed system involves three main steps. In the first step, a filtering feature selection method is applied to select the best features. In the second step, an ad hoc data-driven method is utilized to generate the preliminary fuzzy inference system. Finally, a hybrid optimization method is presented to select the optimum subset of the rules. The study relies on 1,811 samples to evaluate the diagnosis performance of the proposed system. The obtained classification accuracy is very promising with regard to other benchmark classification methods including binary logistic regression (LR) and multilayer perceptron neural network (MLP) with the same attributes. The developed system leads to 100 % sensitivity and 84.7 % specificity, while LR and MLP methods statistically come up with lower figures (65, 78.6 and 65 %, 75.8 %), respectively. Now, a fuzzy supportive tool can be potentially taken as an alternative for the current mortality risk assessment system that are applied in coronary surgeries, and are chiefly based on crisp database.

A new scheme was presented in this study for the evaluation of fetal well-being from the cardiotocogram (CTG) recordings using support vector machines (SVM) and the genetic algorithm (GA). CTG recordings consist of fetal heart rate (FHR) and the uterine contraction (UC) signals and are widely used by obstetricians for assessing fetal well-being. Features extracted from normal and pathological FHR and UC signals were used to construct an SVM based classifier. The GA was then used to find the optimal feature subset that maximizes the classification performance of the SVM based normal and pathological CTG classifier. An extensive clinical CTG data, classified by three expert obstetricians, was used to test the performance of the new scheme. It was demonstrated that the new scheme was able to predict the fetal state as normal or pathological with 99.3 % and 100 % accuracy, respectively. The results reveal that, the GA can be used to determine the critical features to be used in evaluating fetal well-being and consequently increase the classification performance. When compared to widely used ANN and ANFIS based methods, the proposed scheme performed considerably better.

Reduction of health care costs is of paramount importance in our time. This paper is a part of the research which proposes an expert hospital decision support system for resource scheduling. The proposed system combines mathematical programming, knowledge base, and database technologies, and what is more, its friendly interface is suitable for any novice user. Operating rooms in hospitals represent big investments and must be utilized efficiently. In this paper, first a mathematical model similar to job shop scheduling models is developed. The model loads surgical cases to operating rooms by maximizing room utilization and minimizing overtime in a multiple operating room setting. Then a prototype expert system which replaces the expertise of the operations research analyst for the model, drives the modelbase, database, and manages the user dialog is developed. Finally, an overview of the sequencing procedures for operations within an operating room is also presented. © 1995 Plenum Publishing Corporation.

Although Web-based personal health records (PHRs) have been widely deployed, the existing ones have limited intelligence. Previously, we introduced expert system technology and Web search technology into the PHR domain and proposed the concept of an intelligent PHR (iPHR). iPHR provides personalized healthcare information to facilitate users' daily activities of living. The current iPHR is passive and follows the pull model of information distribution. This paper introduces triggers and monitoring into iPHR to make iPHR become active. Our idea is to let medical professionals pre-compile triggers and store them in iPHR's knowledge base. Each trigger corresponds to an abnormal event that may have potential medical impact. iPHR keeps collecting, processing, and analyzing the user's medical data from various sources such as wearable sensors. Whenever an abnormal event is detected from the user's medical data, the corresponding trigger fires and the related personalized healthcare information is pushed to the user using natural language generation technology, expert system technology, and Web search technology.

Web-based personal health records (PHRs) are under massive deployment. To improve PHR's capability and usability, we previously proposed the concept of intelligent PHR (iPHR). By introducing and extending expert system technology and Web search technology into the PHR domain, iPHR can automatically provide users with personalized healthcare information to facilitate their daily activities of living. Our iPHR system currently provides three functions: guided search for disease information, recommendation of home nursing activities, and recommendation of home medical products. This paper discusses our experience with iPHR as well as the open issues, including both enhancements to the existing functions and potential new functions. We outline some preliminary solutions, whereas a main purpose of this paper is to stimulate future research work in the area of consumer health informatics.

Differential diagnosis of renal masses is an important and difficult process. A renal mass diagnostic system (RMDS) developed by using the ILIAD expert system shell has been created for diagnostic consultations and patient simulations. Seventy-two cases of renal mass have been tested on this system and the diagnostic accuracy was compared to that of residents. The overall diagnostic accuracy (75%) for renal masses is significantly better than second-year urological residents (60%) and not worse than urological chief residents (71%). The expert system also displays the cost of the diagnostic procedures so that the user can choose the most cost-effective diagnostic process. We conclude that this powerful renal mass diagnosis system developed by using ILIAD system shell can be used as a teaching, self-training and clinical tool for urological residents.

This paper describes an automated medical record designed to allow providers to enter patient data at the point of care. The system runs on PCs and Macintoshes and uses a graphical user interface and object-oriented programming to take advantage of current mouse and pen technologies. The provider acquires all relevant patient data by pointing and clicking at selections on input screens, many of which contain anatomical drawings to help the provider quickly and accurately describe patient findings. The system also generates a grammatically correct progress note using the problem-oriented structure. Furthermore, items identified in the assessment and plans portion of the program can be ported to expert systems for medical decisions assistance or to billing systems. The system allows the provider to obtain the necessary information on a focused patient visit in less than 5 min or to enter a complete history and physical.

In this paper, we present a three-stage expert system based on a hybrid support vector machines (SVM) approach to diagnose thyroid disease. Focusing on feature selection, the first stage aims at constructing diverse feature subsets with different discriminative capability. Switching from feature selection to model construction, in the second stage, the obtained feature subsets are fed into the designed SVM classifier for training an optimal predictor model whose parameters are optimized by particle swarm optimization (PSO). Finally, the obtained optimal SVM model proceeds to perform the thyroid disease diagnosis tasks using the most discriminative feature subset and the optimal parameters. The effectiveness of the proposed expert system (FS-PSO-SVM) has been rigorously evaluated against the thyroid disease dataset, which is commonly used among researchers who use machine learning methods for thyroid disease diagnosis. The proposed system has been compared with two other related methods including the SVM based on the Grid search technique (Grid-SVM) and the SVM based on Grid search and principle component analysis (PCA-Grid-SVM) in terms of their classification accuracy. Experimental results demonstrate that FS-PSO-SVM significantly outperforms the other ones. In addition, Compared to the existing methods in previous studies, the proposed system has achieved the highest classification accuracy reported so far by 10-fold cross-validation (CV) method, with the mean accuracy of 97.49% and with the maximum accuracy of 98.59%. Promisingly, the proposed FS-PSO-SVM expert system might serve as a new candidate of powerful tools for diagnosing thyroid disease with excellent performance.

Pulmonary Function Tests (PFTs) are very important in the medical evaluation of patients suffering from “shortness of breath”, and they are effectively used for the diagnosis of pulmonary diseases, such as COPD (i.e. chronic obstructive pulmonary diseases). Measurement of Forced Vital Capacity (FVC) and Forced Expiratory Flow in the 1st second (FEV1) are very important for controlling the treatment of COPD. During PFTs, some difficulties are encountered which complicate the comparison of produced graphs with the standards. These mainly include the reluctance of the patients to co-operate and the physicians’ weaknesses to make healthy interpretations. Main tools of the diagnostic process are the symptoms, laboratory tests or measurements and the medical history of the patient. However, quite frequently, most of the medical information obtained from the patient is uncertain, exaggerated or ignored, incomplete or inconsistent. Fuzziness encountered during PFT is very important. In this study, the purpose is to use “fuzzy logic” approach to facilitate reliable and fast interpretation of PFT graphical outputs. A comparison is made between this approach and methodologies adopted in previous studies. Mathematical models and their coefficients for the spirometric plots are introduced as fuzzy numbers. Firstly, a set of rules for categorizing coefficients of mathematical models obtained. Then, a fuzzy rule-base for a medical inference engine is constructed and a diagnostic “expert system COPDes” designed. This program, COPDes helps for diagnosing the degree of COPD for the patient under test.

While artificial brains are in the realm of science fiction, artificial neural networks (ANNs) are scientific facts. An artificial neural network is a computational structure modeled somewhat on the neural structure of the brain; both have many highly interconnected processing elements. These biologically inspired processing elements are taught by feeding examples until the results are acceptable. In the past 5 years, neural networks have become successful in providing meaningful second opinions in clinical diagnosis. In our research, a prototype artificial neural network was trained on numeral ultrasound data of 52 actual cases and then correctly identified renal cell carcinoma from renal cysts and other conditions without diagnostic errors. Our nonlinear artificial neural network was trained on software using the standard backpropagation paradigm on a 80386 microcomputer. Our ANN learned from ultrasound data in 52 cases (17 malignant, 30 cysts, and 5 other) at a Memphis hospital. The trained prototype performed without error on 47 cases which were not in the data used for training. This prototype must be validated by extending this study to more cases.

Iliad is a computerized, expert system for internal medical diagnosis. The system is designed to teach diagnostic skills by means of simulated patient case presentations. We report the results of a controlled trial in which junior students were randomly assigned to received Iliad training on one of two different simulated case mixes. Each group was subsequently tested in both their "trained" and "untrained" case domain. The testing consisted of computerized, simulated patient cases for which no training feedback was provided. Outcome variables were designed to measure the students' performance on these test cases. The results indicate that students made fewer diagnostic errors and more conclusively confirmed their diagnostic hypotheses when they were tested in their trained domain. We conclude that expert systems such as Iliad can effectively teach diagnostic skills by supplementing trainees' actual case experience with computerized simulations.

In a previous paper, the authors described three paradigms applicable to automated medical text analysis. In this paper, the relative importance of the three paradigms is discussed, viz. the relative value of the linguistic word categorization, the semantic paradigm, and the medical fact delineation. The strategy adopted was to limit the linguistic disambiguation and apply probabilistic rules, in order to speed up the analytic process.

In this study, the areas affected from obesity were examined by classifying divergent arteries and body mass index (BMI) of 30 healthy persons and 52 obese persons by using expert systems, and the classifying performances of NEFCLASS and CANFIS, which are expert systems were compared. As a result of this comparison, it is observed that the classifying performance of NEFCLASS is better than that of CANFIS, and the causes of this are examined. Furthermore, it is observed that after these classifications, obesity affects the BMI rather than divergent arteries.

Iliad 4.0 and QMR 2.03 are computer-based diagnostic knowledge bases that can play many roles in decision support and other areas of medical practice, but neither appears ready to assume the role of an expert diagnostic consultant. In contrast to human experts, these programs have problems related to recognition of their own limitations, interpretation of continuous data, recognition of dependent findings, selection of tests, and description of the impact of certain tests. Suggestions to improve these aspects of knowledge bases are offered.

A model for the design of an intelligent tutoring system is presented using artificial intelligence techniques and cognitive processing theories. The model of cognitive processing known as the Knowledge Acquisition and Recall Theory is derived from Anderson's Adaptive Control of Thought Theory. The model is used as a basis for a methodology for the design of an intelligent tutoring system that teaches problem-solving strategies for blood grouping discrepancies. The system developed was tested to determine the effectiveness of the methodology developed and to provide support for the concepts in the model of cognition developed. The preliminary results provide some evidence that an intelligent tutoring system designed using the methodology developed may aid in the knowledge acquisition process. (knowledge acquisition, artificial intelligence, computer-assisted instruction, cognition, intelligent tutoring system).

Group Decision Support Systems (GDSS) are defined and discussed. A GDSS model developed by the author is reviewed in depth for communication of the concepts of GDSS. The model's components are related to health care applications. Questions about unique requirements and level of sophistication in health care applications are explored. What are the differences? What is needed in GDSS software? How do implementation strategies differ? The purpose of this paper is to define and discuss the uniqueness and level of sophistication of GDSS applications in health care. The information requirements and level of information abstraction are the major forces considered in the design of specific medical GDSSs. Data for the GDSS and queries originate both internally and externally to the system. Raw data may be in image form and require extensive analysis by the decision makers for information to be extracted from the raw data. Efforts also are made to relate financial and medical data for better business decisions. This integration often has limited success. Additionally, financial data represent multiple sources and present concerns of validity and reliability. In medical diagnoses the knowledge bases are large and contain thousands of rules. Treatment planning and progress reporting rely on medical records that contain thousands of information items and that often require interpretation by an expert. These information attributes go beyond qualitative versus quantitative definitions and are the author's basis for the analysis presented in this paper.

NEONATE is a prototype of an expert application for the HELP Hospital Information System. Its goal is to improve documentation in the Newborn Intensive Care Unit. The decision module of NEONATE is designed to produce an admission problem list. In this paper, the admission problem list that NEONATE generates was compared to the admission problem list of the current CETUS system for 30 patients. These were compared to a retrospectively constructed gold standard problem list. Of 101 problems in the gold standard list, 56 were on the current admission reports; 82 were found by NEONATE. NEONATE found 31 problems missed on the current admission reports; the current admission reports contained 5 problems missed by NEONATE. The current admission reports contained 9 false positives; whereas NEONATE's reports contained 27. Of the 27, 16 were caused by a single rule in NEONATE. We conclude that an expert system has great potential for improving the documentation of the patient problem list.

In this paper a Maintenance Management Expert System (MMES) is developed. This system is suitable for hospitals as well as for manufacturing organizations. This expert system consists of five sections. These are work order management, equipment management, craftsmen management, material and supply control, and monitoring of maintenance activities. In addition the MMES provides several reports that can be utilized for improving maintenance effectiveness.

LAB VALUES is an expert system that diagnoses the likelihood of various diseases, given the results of routine blood tests. Its knowledge structure employs bit maps. The similarity between a new case and a bit map in the knowledge base is calculated via a similarity score. The system was tested against published cases with definite diagnostic endpoints, and was found to perform well when compared to these published cases. System performance was not improved by incorporating disease frequency into the results.

AI/GEN is an expert model of the diagnosis of deaf-blind syndromes that uses the EXPERT system developed by Rutgers University. Its knowledge structure employs criteria tables for diagnosis of the three types of CHARGE syndrome. The system has been used to test the published diagnostic criteria against the revised expert criteria, the latter being significantly more accurate than the former. The two sets of criteria are also compared with respect to the specificity and sensitivity of diagnosis. Expert systems can be of direct use to experts in refining and revising their diagnostic criteria.

Requirements specification is a key element in a DSS development project because it not only determines what is to be done, it also drives the evolution process. A procedure for requirements elicitation is described that is based on the decomposition of the DSS design task into a number of functions, subfunctions, and operators. It is postulated that the procedure facilitates the building of a DSS that is complete and integrates MIS, modelling and expert system components. Some examples given are drawn from the health administration field.

The object-oriented computing model of Smalltalk/V proved to be very well suited for the creation of a medical expert system on the diagnosis of calcium-phosphorus metabolism abnormalities. A major reason for this was that a disease could be viewed as a self-contained, active, entity. The expert system is implemented in ALEX, an expert system shell that is written in Smalltalk/V. The shell consists of classes and methods that can be modified, as needed, by the developer. The expert system is, at present, 140,000 bytes in size, and includes 43 diseases, arranged in 7 contexts. Sixty case reports were used to test the expert system. In 30 cases, there was complete agreement between clinician and expert system; in 12 cases, the expert system responded appropriately when presented with diseases that were not known to the system; and in 18 cases, the system gave plausible results.

In developing countries, blindness is a major health problem whose control depends on the application of simple measures by frontline workers because, in many of these countries, specialist medical care is not readily available. To assist primary health workers in the management of common and potentially blinding eye disorders, we have developed a prototype computer program for a hand-held computer that incorporates a set of guidelines for diagnosis and treatment. This eye treatment program will help evaluate the potential of such devices for improving health care delivery in developing countries. This approach is now feasible because of recent advances in expert system and portable computer technology.

We describe a software module in an expert system RED, which interprets data related to red cell antibody identification. There are three portions to this module: (1) the problem-solving component, which incorporates the knowledge required for antibody identification as a hierarchy of programs. The programs in the hierarchy organize within themselves small pieces of knowledge represented in the form of production rules, which are capable of making judgments concerning a specific hypothesis; (2) an intelligent data base for storage of patient data, red cell attributes, and test results; (3) the “overview critic” portion, which combines the atomic hypotheses judged favorably by the antibody programs into a unified judgment concerning the case. Overview makes the decision to terminate processing with a conclusion about which antibodies are actually present and what specific further tests need to be performed to resolve any remaining ambiguities.

An experiment is described in which a significant part of the INTERNIST knowledge base for diagnosis in internal medicine is translated into an EXPERT model. INTERNIST employs the largest and broadest knowledge base of all the medical consultation systems that have been developed in recent years. EXPERT is a general system for designing consultation models. The translated model shows reasonable competence in the final diagnostic classification of 431 test cases. There are differences in the internal representation and reasoning strategies of the two systems. However, when a knowledge base has been encoded in a relatively uniform manner, this experiment demonstrates the feasibility of transfer of knowledge between large-scale expert systems.