Abstract
The zoonoses and the infectious and parasitic diseases common to man and animals present a difficult diagnostic challenge for health professionals. Many of the over 200 diseases which fall into this classification are rare, mimic more common illnesses, and are sparsely covered in the medical literature. As a result, they are often misdiagnosed and patients may wait weeks, months, or even years for a definitive diagnosis and appropriate treatment. To address this problem, a computer-based system and a knowledge engineering methodology has been developed which can be used as a diagnostic screening test and a medical knowledge management system for this class of diseases. The system will accept clinical signs and symptoms, occupational data, travel history, animal and vector exposure history, food and water consumption data, results of laboratory tests and diagnostic procedures and will generate a comprehensive differential diagnosis list of possible diseases using simple pattern-matching techniques. In addition, the system enables the user to build a diagnostic plan and to gain quick access to concise, engineered clinical information regarding each disease process. An historical prospective intervention study was performed using actual cases provided by the Texas Department of Health to evaluate the system. The computer intervention significantly decreased the time from first patient visit to the time of suspecting the correct diagnosis for murine typhus (p=.001) and brucellosis (p=. 0001). This demonstrates that the system has the potential to be a sensitive screening mechanism. However, in order to test the specificity, field trials which include a representative sampling of patients and illnesses will have to be carried out. It is hoped that this research will result in a prototype which can then be used to develop other medical diagnostic and reference systems.
Carter, Craig Nash (1993). A computer-based zoonoses differential diagnosis and reference system for the health professions. Texas A&M University. Texas A&M University. Libraries. Available electronically from
https : / /hdl .handle .net /1969 .1 /DISSERTATIONS -1477316.