Medical Diagnosis Process Research Articles

Probabilistic Case Detection for Disease Surveillance Using Data in Electronic Medical Records

This paper describes a probabilistic case detection system (CDS) that uses a Bayesian network model of medical diagnosis and natural language processing to compute the posterior probability of influenza and influenza-like illness from emergency department dictated notes and laboratory results. The diagnostic accuracy of CDS for these conditions, as measured by the area under the ROC curve, was 0.97, and the overall accuracy for NLP employed in CDS was 0.91.

A Framework for Intelligent Medical Diagnosis Using Rough Set with Formal Concept Analysis

Medical diagnosis process vary in the degree to which they attempt to deal with different complicating aspects of diagnosis such as relative importance of symptoms, varied symptom pattern and the relation between diseases them selves. Based on decision theory, in the past many mathematical models such as crisp set, probability distribution, fuzzy set, intuitionistic fuzzy set were developed to deal with complicating aspects of diagnosis. But, many such models are failed to include important aspects of the expert decisions. Therefore, an effort has been made to process inconsistencies in data being considered by Pawlak with the introduction of rough set theory. Though rough set has major advantages over the other methods, but it generates too many rules that create many difficulties while taking decisions. Therefore, it is essential to minimize the decision rules. In this paper, we use two processes such as pre process and post process to mine suitable rules and to explore the relationship among the attributes. In pre process we use rough set theory to mine suitable rules, whereas in post process we use formal concept analysis from these suitable rules to explore better knowledge and most important factors affecting the decision making.

A WT-Based Edge Detection with Adaptive Threshold

The existing edge detection algorithms with wavelet transform need to artificially set the threshold value and are lack of flexibility.To salve the limitations, in this paper, we propose a WT(wavelet transform)-based edge detection algorithm with adaptive threshold, which uses threshold value iteration method to achieve adaptive threshold setting. Comparison of experiment results for the CT image shows that the method which improve the clarity and continuity of the image edge can effectively distinguish edge and noise, and get more completely information of the edge. It has good application value in the fields of medical clinical diagnosis and image processing.

Medicolegal significance of asymmetrical hearing loss in cases of industrial noise exposure

In Australia, the current guidelines for evaluation of noise-induced hearing loss suggest that, in cases of asymmetrical loss, 'the worse ear be equated to the better ear' for purposes of compensation. This study aimed to establish that such a method was prejudicial to the worker (i.e. the plaintiff). In consideration of the legal duty 'to co-construct the ideas of truth and the ideas of justice in the context of legal proceedings', our study objectives were (1) to document the incidence of asymmetrical hearing loss in compensation cases seen in our practice, and (2) to provide a reasoned argument for inclusion of the same for compensation considerations. Open, retrospective, clinical study. Australian plaintiffs with asymmetrical hearing loss (who comprise a significant percentage of industrial hearing loss legal cases) may be excluded from full consideration of their hearing loss as a result of the current guidelines. In contrast to the process of medical diagnosis and treatment, it appeared that the application of accepted probability standards within the legal process may permit inclusion of such clients' hearing loss in compensation considerations. This study included 208 consecutive clients referred by legal practitioners for assessment of hearing loss for compensation purposes. A total of 22.6 per cent of clients (47 of 208) had asymmetrical hearing loss, with the left side having the greater loss in 60 per cent of cases. We believe that asymmetrical hearing loss should be included in compensation considerations, both on medical and legal grounds.

Impact of JPEG 2000 compression on lesion detection in MR imaging

Lossy compression algorithms enable the efficient transmission of large medical image datasets over bandwidth-limited digital networks or facilitate the long-term storage of the daily image production. Although JPEG 2000 has been adopted by DICOM as the standard for lossy-to-lossless compression, still a set of guidelines needs to be derived that allows for the usage of the lossy mode, potentially jeopardizing the accuracy of lesion detection and characterization, and so, of the resulting diagnosis task effectiveness in the medical diagnosis process. In this article the authors present and evaluate a generic methodology for the determination of the minimal bit rate that still ensures an accurate detection in magnetic resonance images of specific lesions. In this article, they demonstrate the methodology for two particular pathologies, i.e., multiple sclerosis and Virchow-Robin space enlargements. Involving qualified personnel, the minimal bit rate is estimated from ROC experiments initially simulated through mathematical observers that are designed with several objective metrics. The mathematical observer models included three variants of the Hotelling observer plus the non-prewhitening matched filter with eye model, while the objective measures are based on distance, correlation, singular value decomposition, and structural similarity. The results indicate that the highest compression without seriously affecting the detection of the studied lesions is achieved at a bit rate of 0.125 bpp. At this value, the detection effectiveness exceeded 95% with less than 5% standard deviation, while only 4.4% of the outcomes were classified as false negatives by the experts and 11.6% as false positives. The optimal bit rate found assures that important information on the small investigated structures is still preserved for their accurate detection and their a posteriori characterization.

LEAD: A Methodology for Learning Efficient Approaches to Medical Diagnosis

Determining the most efficient use of diagnostic tests is one of the complex issues facing medical practitioners. With the soaring cost of healthcare, particularly in the US, there is a critical need for cutting costs of diagnostic tests, while achieving a higher level of diagnostic accuracy. This paper develops a learning based methodology that, based on patient information, recommends test(s) that optimize a suitable measure of diagnostic performance. A comprehensive performance measure is developed that accounts for the costs of testing, morbidity, and mortality associated with the tests, and time taken to reach diagnosis. The performance measure also accounts for the diagnostic ability of the tests. The methodology combines tools from the fields of data mining (rough set theory, in particular), utility theory, Markov decision processes (MDP), and reinforcement learning (RL). The rough set theory is used in extracting diagnostic information in the form of rules from the medical databases. Utility theory is used in bringing various nonhomogenous performance measures into one cost based measure. An MDP model together with an RL algorithm facilitates obtaining efficient testing strategies. The methodology is implemented on a sample problem of diagnosing solitary pulmonary nodule (SPN). The results obtained are compared with those from four alternative testing strategies. Our methodology holds significant promise to improve the process of medical diagnosis.

Pathology raw data in nonclinical laboratory studies for the pharmaceutical industry: The pathologists' view

AbstractPost‐mortem pathology data from nonclinical laboratory studies contain critical information for characterizing a new drug's safety and are pivotal for gaining an appreciation of potential hazards to humans. An understanding of the historical regulatory and scientific perspective on the nature of pathology data is valuable for assuring appropriate and useful audits of pathology material within nonclinical studies. This discussion reviews the specialty of pathology, the process of medical diagnosis, the definitions of ‘raw data’ as they pertain to histopathology data, and the rationale behind the consensus decisions made by industry and the Food and Drug Administration (FDA) in interpreting good laboratory practice (GLP) regulations for this type of data. Copyright © 2004 John Wiley & Sons, Ltd.

Construction of an image recognition process on the basis of optical Fourier diffractometry: application to degraded bone structure classification

A hybrid optical and digital image recognition method (based on optical Fourier diffractometry) applied to the detection of disease-induced pathological changes in bones is presented. X-ray bone images are analyzed in a new type of computer-controlled diffractometer. The set of image parameters, extracted from the diffractogram, is evaluated by statistical analysis. The synthetic statistical image descriptors, constructed based on the groups of training images, enable recognition of bone samples with degraded bone structure and then recognition of disease. About 89% of images were classified correctly. After optimization, this method can be applied to a computer-aided medical diagnosis process.

Drinking, Driving, and Deterrence: The Effectiveness and Social Costs of Alternative Policies

TRAFFIC accidents are one of the leading causes of death in the United States and are the leading cause of death for people aged three to thirtyfour. It is estimated that alcohol is involved in approximately half of all fatal accidents.' The deaths and injuries of nondrinking drivers, passengers, and pedestrians both arouse public opinion and provide a clear-cut efficiency rationale for policies to correct the negative externalities.2 There are two types of policies aimed at reducing drunk driving. The first type of law attempts to deter drunk driving. The second type of law indirectly reduces drunk driving by controlling alcohol consumption, either in general or in high-risk groups. Examples of alcohol control policies

Cooperative decision support for medical diagnosis

In this paper, the issue of group-based cooperative decision support is examined. The notion of a group as an essential element for management in a cooperative context is discussed. Viewpoints, or interfaces that are tailored to the requirements of user tasks, are explained, and group viewpoints are introduced. An analysis of what groups participate in the hospital management task is illustrated, and task-specific viewpoints for these groups are established. Examples of two decision dialogues, the first between members of a group, and the second between members of different groups, are presented. First, the environmental elements of cooperative decision-making are established. Then interface elements that are required to support cooperative decision-making are put in place. The phases of a computer-mediated group decision are then outlined. The process of cooperative medical diagnosis is then considered in the light of the theoretical development, and the specific interface of the medical environment is established. Finally, a review of the adequacy of the entire approach is conducted, and a possible implementation platform is considered.

New applications of CdTe nuclear detectors

The unique properties of CdTe nuclear detectors offer a number of significant benefits to researchers, clinicians and engineers who have special requirements relating to size, sensitivity, and operating temperature. The use of CdTe nuclear detectors is continuing to expand into new areas as its unique properties and capabilities become better known and as improved manufacturing techniques increase the overall quality and yield of detectors. There is also continuing growth in areas where CdTe has become established, such as include biomedical research, medical diagnosis, cancer therapy, medical and industrial imaging and process control. In particular, there have been many new procedures developed in the field of nuclear medicine, many of which require a small, sensitive detector. This paper reviews several new areas for CdTe detectors and detector arrays.

The complementary process of fuzzy medical diagnosis and its properties

The composite fuzzy relational equation and its applications to the medical diagnosis were first proposed by Sanchez. He also gave the method to derive the medical knowledge from the clinical data, which consist of two fuzzy relations on a set of patients and a set of propositions that represent symptoms or diagnoses. His method, however, gives attention to only one view of these data. In this paper the complementary process to Sanchez's fuzzy diagnosis process is introduced by reinterpreting the clinical data on the domains of negated propositions. Some properties of the fuzzy diagnosis processes are also given.

Clinical judgment, expert programs, and cognitive style: a counter-essay in the logic of diagnosis.

The question of the extent to which one can rationally reconstruct the process of medical diagnosis and reduce it to an algorithm is explored. The act of diagnostic insight is such that a computational program cannot 'catch on' in the way that a competent diagnostician can. Clinical diagnostic reasoning in a particular case requires as a necessary condition an extraordinarily complex and rich structure of background knowledge as well as an intuitive element, such as is manifest when one 'catches on' to a joke. Computers cannot 'catch on' due to their limitations in the computational mode. As a consequence, there can be no computer simulation of an essential element in diagnostic procedures.

Medical Image Diagnosis and Image Processing System

Medical Image Diagnosis and Image Processing System

The psychosomatic approach and the problem of diagnosis

This paper examines some of the characteristics and aims of the process of medical diagnosis and the problems posed by the seeming opposition between disease-centered and person-centered diagnosis. The latter was one of the basic tenets of the psychosomatic approach from its inception. Its application to medical practice, however, has been hindered by the lack of an appropriate theoretical framework which might help avoid the one-sidedness prevailing in the discourse of different medical specialties. The building up of an integrative medicine has proceeded on the basis of addition and not on the basis of organic development from a theory-derived set of principles. In order to circumvent the problems associated with such state of affairs, it is proposed that multiaxial systems of diagnosis, like the ones currently in use in psychiatry, be adopted. These should be improved and expanded in order to take into account additional dimensions and data sources not hitherto taken into consideration.

The Process of Medical Diagnosis: Routes of Mathematical Investigations

This paper is an analysis of the most important mathematical aspects of medical diagnosis: logical probability, rationality and decision theory, gambling models, pattern analysis, hazy and fuzzy subsets theory and, finally, the stochastic inquiry process.

A Markovian Decision Model for Clinical Diagnosis and Treatment Applied to the Respiratory System

The complex process of medical diagnosis has traditionally relied on the experience and judgement of the clinician. With the increased application of systems ideas to medical problems in general, it is timely to consider their application to the diagnostic situation. Decision theory is well established, and several decision models have been applied to medical problems. A unified approach to clinical decision making is presented. This approach combines partially observable Markov decision processes with cause-effect models as a probabilistic representation of the diagnostic process. This new class of model has a potential application to medical diagnosis and treatment, and a respiratory system example is presented. The methodology is given for combining the patient states of health, the clinician's state of knowledge of the cause-effect representation from the observation space, and finally the treatment decisions with which to restore the patient to a more desirable state of health.

Practical problems in the use of computers in medical diagnosis

For the computer to assist in the medical diagnosis and treatment processes, a series of systematic mathematical algorithms must be developed that correspond in some sense to the reasoning used by the diagnostician. In terms of the propositional calculus of symbolic logic, the patient presents a symptom profile, G, which is a Boolean function of possible symptoms, either present or absent or undetermined. The diagnosis, which is also a Boolean function, f, of possible diseases, is found from the fundamental formula of medical diagnosis, E → (G → f), where E is the Boolean function representing medical knowledge. An example of an elementary computation is given to illustrate the process of solving the fundamental formula. In cases where complete symptom information is unavailable, the computer can apply Bayes' formula, which correlates the conditional probability of having a disease complex given a certain symptom profile with that of having the given symptom profile given the disease complex and the total probability of the patient's having the disease complex. A simple illustrative example is given of the application of Bayes' formula. An uncertainty principle exists in the collection of statistics for determining the total probability, due to the time lag in processing data. A method is given for handling data from patients presenting incomplete symptom profiles. When medical knowledge is so voluminous as to make impractical the recording of all possible symptom-disease combinations, then techniques developed in pattern-recognition studies can be utilized. The least-probable symptom complex will be omitted, and the loss can be compensated for by comparing the patient's given symptom profile with weighted possible symptom complexes. Again, an illustrative example is given. After the probabilities for possible alternative diagnoses have been determined, the treatment plan must be determined. We distinguish three basic kinds of problem: 1) treatment decisions under certainty, which involve simple value-theory considerations and linear programming; 2) treatment decisions under risk, where the probabilities of alternative diagnoses are known, which involve optimization of a mathematical expectation; and 3) treatment decisions under uncertainty, where the probabilities of the alternative diagnoses are unknown, which are amenable to the application of game theory. Such decisions all involve value determinations, which can be broken down into tangible and intangible values. When alternative diagnoses remain, the response of the patient to the treatment chosen is a further symptom, which can be used in reevaluating the diagnosis. In such cases the techniques of dynamic programming may be applied. The problem of determining the transition probabilities of a patient's moving from one state of health to another during a certain course of treatment presents difficulties, and a method for alleviating the problem is presented, utilizing Bayes' formula to process current patient charts instead of relying on past record searches. Conditional probabilities can be further utilized in the comparison of two treatments. Finally, a brief survey is given of applications of computer-diagnosis aids as reported in the literature.