Suppr超能文献

[肺功能测试中的医学实践支持系统]

[Medical practice-supporting system in pulmonary function testing].

作者信息

Kino H

机构信息

Mitsui Memorial Hospital, Tokyo.

出版信息

Rinsho Byori. 1997 May;45(5):438-41.

PMID:9170971
Abstract

A pulmonary function testing system which automates measurement, storage and reporting as much possible has been functioning in Mitsui Memorial Hospital since 1992. Data obtained from four measurement apparatus are transferred to the central microcomputer by a local area network (LAN). To facilitate automatic interpretation, we developed a knowledge-based system using 87 production rules called 'PADRES', because the knowledge-based system had the following advantages over other systems; (1) The expression of knowledge is understandable and new pieces of knowledge can be easily added, (2) the process of reasoning can be clearly shown, and (3) the accumulation of many instances of data is not required. Using 'PADRES', it takes 2.6 seconds to interpret one set of data on average. Our system performs about 5,000 routine pulmonary function tests per year without faults. Because the program for interpretation is a separate module written in C language, it is readily portable to other operating systems (OS), although its original target was OS-9, a multitask operating system for microcomputers. In conclusion, the application of a knowledge-based system for microcomputers in the pulmonary function test laboratory was considered successful.

摘要

自1992年以来,一个尽可能实现测量、存储和报告自动化的肺功能测试系统一直在三井纪念医院运行。从四台测量仪器获取的数据通过局域网(LAN)传输到中央微型计算机。为便于自动解读,我们开发了一个基于知识的系统,使用87条称为“PADRES”的产生式规则,因为该基于知识的系统相对于其他系统具有以下优势:(1)知识表达易于理解,新的知识片段可以轻松添加;(2)推理过程可以清晰显示;(3)不需要积累大量数据实例。使用“PADRES”,平均解释一组数据需要2.6秒。我们的系统每年无故障地执行约5000次常规肺功能测试。由于解释程序是用C语言编写的一个单独模块,尽管其最初目标是用于微型计算机的多任务操作系统OS-9,但它很容易移植到其他操作系统(OS)。总之,基于知识的系统在微型计算机上应用于肺功能测试实验室被认为是成功的。

相似文献

6
Knowledge-based automation.基于知识的自动化
Clin Lab Med. 1988 Dec;8(4):713-22.

文献AI研究员

20分钟写一篇综述,助力文献阅读效率提升50倍。

立即体验

用中文搜PubMed

大模型驱动的PubMed中文搜索引擎

马上搜索

文档翻译

学术文献翻译模型,支持多种主流文档格式。

立即体验