利用機器學習根據質譜資料建立細菌株快速藥敏預測與自動化分析平台;An Automatic Analysis Platform for Rapid Antibiotics Susceptibility Prediction to Bacteria Based on Mass Spectra Data Using a Machine Learning Approach

NCU Institutional Repository > 資訊電機學院 > 資訊工程學系 > 研究計畫 > Item 987654321/82305

jsp.display-item.identifier=請使用永久網址來引用或連結此文件: http://ir.lib.ncu.edu.tw/handle/987654321/82305

题名:	利用機器學習根據質譜資料建立細菌株快速藥敏預測與自動化分析平台;An Automatic Analysis Platform for Rapid Antibiotics Susceptibility Prediction to Bacteria Based on Mass Spectra Data Using a Machine Learning Approach
作者:	洪炯宗;盧章智
贡献者:	國立中央大學資訊工程學系
关键词:	藥敏檢驗;基質輔助激光解吸電離-飛行時間質譜;機器學習;多重抗藥性;antibiotics susceptibility test;matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF MS);machine learning;multidrug resistance
日期:	2020-01-13
上传时间:	2020-01-13 14:39:18 (UTC+8)
出版者:	科技部
摘要:	藥敏檢驗為體外測試微生物對於藥物的反應（「具抗藥性」或「不具抗藥性」），以提供臨床醫師抗生素使用之指引。傳統微生物檢驗的方法確認細菌之藥敏反應需要數天之時間，導致無法於第一時間給予最正確之給藥。相對，快速藥敏主要目標為即早並正確給藥，進而達到減低死亡率、避免抗藥性，以及縮短住院天數之效益。近年來，基質輔助激光解吸電離-飛行時間質譜（MALDI-TOF MS）已被廣泛應用於臨床微生物檢驗，經由不同菌種之質譜些微差異，即可得到高度準確性之菌種辨認。然而，現行方法尚無法根據質譜之些微差異，以高精準度地預測抗藥性。本計畫之目的為利用質譜資料建立模型，以預測細菌之藥敏反應。其中，此計畫之資料為林口長庚醫院檢驗醫學科多年來蒐集臨床之細菌質譜與其藥敏反應結果，本計畫首先建立一資料庫系統儲存質譜與藥敏資料，再經由機器學習與適當之特徵挑選方法建立藥敏反應模型。最後將進一步尋找關鍵峰值特徵，進行更進一步的微生物實驗找尋對應之蛋白質片段，以探究產生抗藥之原因，提供製藥之參考。最後，將應用於實際臨床醫學場域，提供最即時之預測以即早正確給藥。 ;Antibiotics susceptibility test (AST) is an in-vitro test for providing information of microorganism against antibiotics. The test result could be susceptible or resistant, which guides clinical physicians for correct use of antibiotics. The current AST procedure in clinical microbiology laboratory would spend several days, which hinder correct and timely treatment against infectious disease. In contrast, rapid AST aims to provide accurate AST in shorter turn-around-time, which could reduce mortality, avoid drug resistance, and shorten length of stay in hospital. In recent years, matrix-assisted laser desorption ionization time of flight mass spectrometry (MALDI-TOF MS) has become an alternative and powerful tool in clinical microbiology laboratories for rapid bacteria species identification. However, rapid AST still cannot be obtained from the MALDI-TOF MS spectra. Therefore, the primary objective of this project is to construct a large database system composed of MALDI-TOF MS data and AST obtained from Chang Gung Memorial Hospitals (Linkou and Kaohsiung branches). On the basis of the database, prediction models capable of providing rapid AST will be developed, validated, and tested by an independent dataset. Moreover, informative features will be selected by feature selection methods to provide clues for further development of new drugs. Finally, a web-based tool for rapid AST will be developed to apply in the clinical medicine.
關聯:	財團法人國家實驗研究院科技政策研究與資訊中心
显示于类别:	[資訊工程學系] 研究計畫

文件中的档案:

档案	描述	大小	格式	浏览次数
index.html		0Kb	HTML	283	检视/开启

在NCUIR中所有的数据项都受到原著作权保护.

社群 sharing

数据加载中.....