Laurikkala J, Juhola M, Penttinen J, Aukee P
Department of Computer Science and Applied Mathematics, University of Kuopio, Finland.
Stud Health Technol Inform. 1997;43 Pt B:671-5.
Female urinary incontinence is a difficult problem for a patient but also for a physician. In the differential diagnosis of female urinary incontinence the physician has to determine a diagnostic class for the patient. This task is complex because of the unreliable patient history and the overlapping class boundaries. In order to develop an expert system to help the physician, a retrospective investigation on the incontinent women was performed to detect the potential expert system parameters. Also a diagnosis table was constructed from the expected values of parameters and the diagnostic classes. The results from K-means cluster analysis indicate that it is possible to develop the expert system on basis of the defined parameters and classes.
女性尿失禁对患者和医生来说都是一个难题。在女性尿失禁的鉴别诊断中，医生必须为患者确定一个诊断类别。由于患者病史不可靠且类别界限重叠，这项任务很复杂。为了开发一个专家系统来帮助医生，对尿失禁女性进行了回顾性调查，以检测潜在的专家系统参数。还根据参数的期望值和诊断类别构建了一个诊断表。K均值聚类分析的结果表明，基于定义的参数和类别开发专家系统是可行的。
