A Rule-Based Disease Diagnostic System Using A Temporal Relationship Model
碩士 === 逢甲大學 === 資訊工程所 === 93 === In many cases, medical diagnostic data are related to each other and are time-dependent that will make the analysis of medical diagnostic data difficult and result in complicated medical decision rules. The patient data including laboratory testing data and recorded...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | en_US |
Published: |
2005
|
Online Access: | http://ndltd.ncl.edu.tw/handle/95076974467306376293 |
id |
ndltd-TW-093FCU05392026 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-093FCU053920262015-10-13T11:20:16Z http://ndltd.ncl.edu.tw/handle/95076974467306376293 A Rule-Based Disease Diagnostic System Using A Temporal Relationship Model 一個使用時間關係模型的法則基準之疾病診斷系統 Chien-Chih Wang 王建智 碩士 逢甲大學 資訊工程所 93 In many cases, medical diagnostic data are related to each other and are time-dependent that will make the analysis of medical diagnostic data difficult and result in complicated medical decision rules. The patient data including laboratory testing data and recorded vital signal readings are timestamped and the consecutive timestamps of a testing data form a time interval. Since the patient data are timestamped, a disease is usually determined by multiple occurrences of abnormal testing data, it is associated with a time interval. In this thesis, we use a temporal model to express temporal relationships among diseases that may have mutual affect potentially. The temporal model defines six types of relationships: before, after, contained, contains, start_overlap, and end_overlap. First, the temporal relationship between diseases of each patient is searched, and then, the temporal relationships of all patients are analyzed to determine the correlation of these diseases. The timestamped diagnostic data are built as facts and the physician specified rules of diseases are built as inference rules of the inference engine. A patient is diagnosed of a disease if his/her patient data satisfy some rules specified by physicians. The temporal relationship model is implemented as a rule based system using the Java based expert system, Jess. The rules for determining disease diagnoses and temporal relationships are written as text files and are input to the system. The system can be used for the other diagnoses and/or diseases by changing the rules without modifying the system program. Hence, the major goal of this thesis is to provide a system kernel for applications of medical diagnosis system. none 黃秋煌 2005 學位論文 ; thesis 42 en_US |
collection |
NDLTD |
language |
en_US |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 逢甲大學 === 資訊工程所 === 93 === In many cases, medical diagnostic data are related to each other and are time-dependent that will make the analysis of medical diagnostic data difficult and result in complicated medical decision rules. The patient data including laboratory testing data and recorded vital signal readings are timestamped and the consecutive timestamps of a testing data form a time interval. Since the patient data are timestamped, a disease is usually determined by multiple occurrences of abnormal testing data, it is associated with a time interval. In this thesis, we use a temporal model to express temporal relationships among diseases that may have mutual affect potentially. The temporal model defines six types of relationships: before, after, contained, contains, start_overlap, and end_overlap. First, the temporal relationship between diseases of each patient is searched, and then, the temporal relationships of all patients are analyzed to determine the correlation of these diseases. The timestamped diagnostic data are built as facts and the physician specified rules of diseases are built as inference rules of the inference engine. A patient is diagnosed of a disease if his/her patient data satisfy some rules specified by physicians. The temporal relationship model is implemented as a rule based system using the Java based expert system, Jess. The rules for determining disease diagnoses and temporal relationships are written as text files and are input to the system. The system can be used for the other diagnoses and/or diseases by changing the rules without modifying the system program. Hence, the major goal of this thesis is to provide a system kernel for applications of medical diagnosis system.
|
author2 |
none |
author_facet |
none Chien-Chih Wang 王建智 |
author |
Chien-Chih Wang 王建智 |
spellingShingle |
Chien-Chih Wang 王建智 A Rule-Based Disease Diagnostic System Using A Temporal Relationship Model |
author_sort |
Chien-Chih Wang |
title |
A Rule-Based Disease Diagnostic System Using A Temporal Relationship Model |
title_short |
A Rule-Based Disease Diagnostic System Using A Temporal Relationship Model |
title_full |
A Rule-Based Disease Diagnostic System Using A Temporal Relationship Model |
title_fullStr |
A Rule-Based Disease Diagnostic System Using A Temporal Relationship Model |
title_full_unstemmed |
A Rule-Based Disease Diagnostic System Using A Temporal Relationship Model |
title_sort |
rule-based disease diagnostic system using a temporal relationship model |
publishDate |
2005 |
url |
http://ndltd.ncl.edu.tw/handle/95076974467306376293 |
work_keys_str_mv |
AT chienchihwang arulebaseddiseasediagnosticsystemusingatemporalrelationshipmodel AT wángjiànzhì arulebaseddiseasediagnosticsystemusingatemporalrelationshipmodel AT chienchihwang yīgèshǐyòngshíjiānguānxìmóxíngdefǎzéjīzhǔnzhījíbìngzhěnduànxìtǒng AT wángjiànzhì yīgèshǐyòngshíjiānguānxìmóxíngdefǎzéjīzhǔnzhījíbìngzhěnduànxìtǒng AT chienchihwang rulebaseddiseasediagnosticsystemusingatemporalrelationshipmodel AT wángjiànzhì rulebaseddiseasediagnosticsystemusingatemporalrelationshipmodel |
_version_ |
1716841037684539392 |