<p> Bayes' formula may be written as follows: </p> <p> P(yᵢ|X) = P(X|yᵢ)・P(yᵢ)/j=K Σ j=1 P(X|yⱼ)・P(yⱼ) where (1) </p> <p> Y = {y₁, y₂,..., y_K} </p> <P> X = {x₁, x₂,..., xₖ} </p> <p> Assuming independence of attributes x₁, x₂,..., xₖ, Bayes' formula may be rewritten as follows: </p> <p> P(yᵢ|X) = P(x₁|yᵢ)・P(x₂|yᵢ)・...・P(xₖ|yᵢ)・P(yᵢ)/j=K Σ j=1 P(x₁|yⱼ)・P(x₂|yⱼ)・...・P(xₖ|yⱼ)・P(yⱼ) (2) </p> <p> In medical diagnosis the y's denote disease states and the x's denote the presence or absence of symptoms. Bayesian inference is applied to medical diagnosis as follows: for an individual with data set X, the predicted diagnosis is the disease yⱼ such that P(yⱼ|X) = max_i P(yᵢ|X), i=1,2,...,K (3) </p> <p> as calculated from (2). </p> <p> Inferences based on (2) and (3) correctly allocate a high proportion of patients (>70%) in studies to date, despite violations of the independence assumption. The aim of this thesis is modest, (i) to demonstrate the applicability of Bayesian inference to the problem of medical diagnosis (ii) to review pertinent literature (iii) to present a Monte Carlo method which simulates the application of Bayes' formula to distinguish among diseases (iv) to present and discuss the results of Monte Carlo experiments which allow statistical statements to be made concerning the accuracy of Bayesian inference when the assumption of independence is violated. </p> <p> The Monte Carlo study considers paired dependence among attributes when Bayes' formula is used to predict diagnoses from among 6 disease categories. A parameter which measured deviations from attribute independence is defined by DH=(j=6 Σ j=1|P(x_B|x_A,yⱼ)-P(x_B|yⱼ)|)/6, where x_A and x_B denote a dependent attribute pair. It was found that the correct number of Bayesian predictions, M, decreases markedly as attributes increasing diverge from independence, ie, as DH increases. However, a simple first order linear model of the form M = B₀+B₁・DH does not consistently explain the variation in M. </p> / Thesis / Master of Science (MSc)

Call number: LD2668 .T4 1957 B38 / Master of Science

M.Tech. (Biomedical Technology) / Please read abstract in the full-text document

The purpose of this study is to develop a technique by which a physician may use a predetermined data base to derive a preliminary diagnosis for a patient with a given set of symptoms. The technique will not yield an absolute diagnosis, but rather will point the way to a set of most likely diseases upon which the physician may concentrate his efforts. There will be no reliance upon a data base compiled from poorly kept medical records with non-standardization of terminology. While this study produces a workable tool for the physician to use in the process of medical diagnosis, the ultimate responsibility for the patient's welfare must still rest with the physician.

Der vorliegende Beitrag betrachtet die Entwicklung der Sonografie und erläutert die heutigen technischen Verfahren sowie deren Anwendung. Die Geschichte der Erkennung des physikalischen Phänomens Schall geht zwar bis in die Antike zurück, die Nutzung zur medizinischen Diagnostik begann jedoch erst in der Mitte des vorigen Jahrhunderts. Die Ultraschalldiagnostik ist ein heute aus dem klinischen Alltag nicht mehr wegzudenkendes nicht-invasives Verfahren zur Überwachung der normalen und gestörten Schwangerschaft. Bereits 1979 wurde das Verfahren in die Routinediagnostik im Rahmen der Schwangerenbetreuung eingeführt. Die technische Entwicklung hat zu einer erheblichen Verbesserung in der Bildqualität und damit zur besseren Erkennbarkeit des Fetus, seiner Plazenta und Hüllen geführt. / The following contribution serves as an introduction to the history of ultrasonography and explains both the current technical methods and their application. First knowledge of the phenomenon of sonic waves can be traced back to ancient times. Its utilisation in medical diagnosis, however, did not start until the middle of the last century. In daily medical routine, one cannot today imagine life without the non-invasive method of ultrasonic diagnostics for the monitoring of regular and abnormal pregnancies. The method was already introduced into routine pregnancy observation in 1979. Technical innovation has remarkably improved image processing, accompanied by unprecedented progress in visualisation of the foetus, its placenta and the amnion membranes.

Forming a medical diagnosis is a complicated reasoning process undertaken by physicians. Although there has been much research focusing on clinical reasoning approaches, there is limited empirical evidence in relation to causal attribution in medical diagnosis. The research on which this thesis is based explored and examined the social process of medical diagnosis and provides an explanation of the clinical reasoning and causal attribution used by physicians. The research was undertaken in an Emergency Department within an acute hospital, the data were collected using mixed method approach including one to one semi-structured interviews with individual physicians; observation of their medical assessments of patients and secondary data analysis of the subsequent recorded medical notes. The study involved 202 patients and 26 physicians. The analysis of the physicians’ semi-structured interviews, shows how physicians describe the diagnostic step process and how they blend their clinical reasoning skills and professional judgment with evidence-based medicine. Physicians apply prior learning of taught biomedical and pathophysiological knowledge to question patients using pattern recognition of common signs and symptoms of disease. These findings are portrayed through taped narratives of the physician/patient interaction during the medical diagnostic process, which shows how physicians control the medical encounter. The analysis/interpretation of documentary evidence (recorded medical notes) provides an insight into the way in which physicians used the information gathered during the diagnostic step process. By using SPSS it was possible to cluster the cases (individual patients) into groups. This stage-ordered classification procedure demonstrated commonality amongst individual cases whilst highlighting the uniqueness of any cases. A pattern emerged of two groups of cases: Group 1 - comprised of patients with the presenting complaints of chest pain, shortness of breath, collapse, abdominal pain, per rectal bleed, nausea, vascular and neurological problems and Group 2 - comprised of patients presenting with trauma, mechanical falls, miscarriage/gynaecological problems, allergies/rashes and dental problems. Findings show that the clinical reasoning approaches used varied according to the complexity of the patient’s presenting complaint. The recorded medical notes for the patients in Group 1, were comprehensive and demonstrated a combined approach of hypothetic-deductive and probabilistic reasoning which enabled the physicians to deal with the degree of uncertainty that is inherent in medicine. The recorded process in the medical notes was shortened for the majority of patients in Group 2, and here the clinical reasoning approach used was found to deterministic. It is acknowledged, that this is not always the case. By using crisp set QCA it was possible to explore causal conditions consistent with Group 1. Further analysis led to examination of the link of causal conditions presented in the medical notes with the individual impression/working diagnosis made by physicians.

No description available.

Thesis(M.S.)--Auburn University, 2005. / Abstract. Vita. Includes bibliographic references.

- / -

- / -