Application of Data Mining Techniques to Assess Postoperative Pain Relief Time
碩士 === 國立中正大學 === 資訊管理系醫療資訊管理研究所 === 104 === Management and relief of pain is the most basic demand of human. Domestic and foreign institutions have set pain as the fifth vital sign and relieving pain of patients is always one of the indicators of health care quality. Pain is the first problem in pa...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2015
|
Online Access: | http://ndltd.ncl.edu.tw/handle/a9bps3 |
id |
ndltd-TW-104CCU00777001 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-104CCU007770012019-05-15T22:17:27Z http://ndltd.ncl.edu.tw/handle/a9bps3 Application of Data Mining Techniques to Assess Postoperative Pain Relief Time 應用資料探勘技術評估術後病人疼痛緩解時間 Wu Mei- Feng 巫美鳳 碩士 國立中正大學 資訊管理系醫療資訊管理研究所 104 Management and relief of pain is the most basic demand of human. Domestic and foreign institutions have set pain as the fifth vital sign and relieving pain of patients is always one of the indicators of health care quality. Pain is the first problem in patients post operation. Previous studies showed 77% to 98% of patients ever experienced post-operational pain; among them, 40% to 50% didn’t receive ideal treatment of pain relief. This study used mining technology to evaluate if pain duration after surgery is less than 3 days. This study recruited patients receiving caesarean at one regional hospital in ChiayiCity of Taiwan. Purposive sampling principle was used to collect data. 639 cases were collected totally and data mining in the supervised classification methods was undertaken to build the prediction model. 10-fold cross-validation method was used to evaluate the accuracy of each prediction mode and the best mode was selected later. The classifier prediction mode constructed with classification tree (J40) showed that for patients after caesarean with pain score less than 4 over one to three days, the correct predictive value was 89.41%, the sensitivity was 90.11% and the specificity was 88.49%. The correct value of prediction with J49 mode was 0.9. This study should be able to assist the team of obstetrics and gynecology to build the appropriate predicting mode of pain relief and cough give the suggestion of effective pain treatment. HuYa-Han 胡雅涵博士 2015 學位論文 ; thesis 50 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立中正大學 === 資訊管理系醫療資訊管理研究所 === 104 === Management and relief of pain is the most basic demand of human. Domestic and foreign institutions have set pain as the fifth vital sign and relieving pain of patients is always one of the indicators of health care quality. Pain is the first problem in patients post operation. Previous studies showed 77% to 98% of patients ever experienced post-operational pain; among them, 40% to 50% didn’t receive ideal treatment of pain relief. This study used mining technology to evaluate if pain duration after surgery is less than 3 days.
This study recruited patients receiving caesarean at one regional hospital in ChiayiCity of Taiwan. Purposive sampling principle was used to collect data. 639 cases were collected totally and data mining in the supervised classification methods was undertaken to build the prediction model. 10-fold cross-validation method was used to evaluate the accuracy of each prediction mode and the best mode was selected later.
The classifier prediction mode constructed with classification tree (J40) showed that for patients after caesarean with pain score less than 4 over one to three days, the correct predictive value was 89.41%, the sensitivity was 90.11% and the specificity was 88.49%. The correct value of prediction with J49 mode was 0.9. This study should be able to assist the team of obstetrics and gynecology to build the appropriate predicting mode of pain relief and cough give the suggestion of effective pain treatment.
|
author2 |
HuYa-Han |
author_facet |
HuYa-Han Wu Mei- Feng 巫美鳳 |
author |
Wu Mei- Feng 巫美鳳 |
spellingShingle |
Wu Mei- Feng 巫美鳳 Application of Data Mining Techniques to Assess Postoperative Pain Relief Time |
author_sort |
Wu Mei- Feng |
title |
Application of Data Mining Techniques to Assess Postoperative Pain Relief Time |
title_short |
Application of Data Mining Techniques to Assess Postoperative Pain Relief Time |
title_full |
Application of Data Mining Techniques to Assess Postoperative Pain Relief Time |
title_fullStr |
Application of Data Mining Techniques to Assess Postoperative Pain Relief Time |
title_full_unstemmed |
Application of Data Mining Techniques to Assess Postoperative Pain Relief Time |
title_sort |
application of data mining techniques to assess postoperative pain relief time |
publishDate |
2015 |
url |
http://ndltd.ncl.edu.tw/handle/a9bps3 |
work_keys_str_mv |
AT wumeifeng applicationofdataminingtechniquestoassesspostoperativepainrelieftime AT wūměifèng applicationofdataminingtechniquestoassesspostoperativepainrelieftime AT wumeifeng yīngyòngzīliàotànkānjìshùpínggūshùhòubìngrénténgtònghuǎnjiěshíjiān AT wūměifèng yīngyòngzīliàotànkānjìshùpínggūshùhòubìngrénténgtònghuǎnjiěshíjiān |
_version_ |
1719128205118406656 |