透過您的圖書館登入
IP:18.116.239.195
  • 學位論文

自動化資料比對法應用在醫學中心放射免疫資料之分析檢測

Application of Automatic Data Matching Method in Assessing the Radioimmunoassay Data in a Medical Center

指導教授 : 王居卿
共同指導教授 : 何錦堂(Giin-Tarng Ho)

摘要


本研究的宗旨在探討醫學中心實驗室資訊系統(Laboratory Information System, LIS) 與醫院資訊系統(Hospital Information System, HIS)子系統之一的科室報到系統(Departmental Registration System, DRS)之間所存在的資料品質問題。本研究從有關LIS、HIS 與DRS 等系統間的介面、資料品質問題以及解決資料品質問題的方法的文獻探討開始著手。在本研究中我們使用EXCEL 軟體之Microsoft Visual Basic 的環境下所撰寫的分析程式來整合LIS 與DRS 間資料異質化問題並且進行自動化的資料比對。該分析程式每次可以在30 秒的時間內自動比對LIS 與DRS 各300 筆資料,其中每一筆資料包括了姓名、病歷號、檢查流水號、以及平均約3 個醫另品項的資料。除了自動比對之外,並且可以自動將問題資料整理成EXCEL 表格資料以供後續分析與校正。 根據本研究範例所顯示,醫院LIS 與DRS 兩個不同系統之間存在著資料異質性的問題,該問題使得人工比對既困難又容易出錯,也使得自動化的資料比對無法進行。透過兩系統的資料格式(包括檢查流水號與醫令名稱)的統一, 本研究所提出的方法析解決了資料異質性的問題。 透過自動化的資料比對,我們可以找出LIS 與DRS 各自資料的重複(包括病歷號、檢查流水號與檢查品項等)、LIS 與DRS 資料的不一致(包括姓名、檢查流水號、檢查品項)。自動化的資料比對可以及時的偵測資料品質的問題,進而避免後續健保申報不實、成本浪費與醫療品質等問題。就管理的角度而言,自動化的資料比對法提供管理階層一個快速、即時、準確的資料品質問題偵測工具,彌補兩個不同系統因為資料異質性所潛藏的資料品質問題。

並列摘要


The aim of this study is to investigate the data quality problems existing in the Laboratory Information System (LIS) and the Departmental Registration System (DRS), which is one of the subsystem of the Hospital Information System (HIS).We start from literature review regarding the interfaces among LIS, HIS and DRS, data quality problems, and solutions to solve data quality problems. We use a customized analysis tool, which is developed in-house using Microsoft Visual Basic on the platform of EXCEL software, to solve the data heterogeneity between LIS and DRS and to perform data matching automatically. By using this tool, we are able to perform data matching between LIS and DRS at a rate of 300 groups of data within 30 seconds, with each group of data comprising patient name, history number, accessing number, and an average of 3 medical items. Besides, the tool also provides automatic summarization of the detected data errors on an independent EXCEL data sheet. According to our study case, we found data heterogeneity between LIS and DRS, making manual data matching difficult and erroneous and automatic data matching impossible. Our data analysis tool integrated the heterogeneous data into uniform data to allow automatic data matching, which was proven capable to detect intra-system data repeatability and inter-system data inconsistency. Automatic data matching provides in-time detection of data quality problems, avoids health insurance declaration problems, cost waste problems, and medical care quality problems, etc. From the viewpoint of management, automatic data matching provides the administrator a rapid, in-time, accurate tool for detecting data quality problem, making up the data quality problem hiding behind the data heterogeneity between different systems.

參考文獻


田聖才 (2000)。 電腦閱卷系統的研究。 測驗統計年刊 ,8:141-164。
Dorneles, C. F., Heuser, C. A., Lima, A. E. N., da Silva, A. S. and de Moura, E. S. (2004). Measuring similarity between collection of values. In WIDM ’04:Proceedings of the sixth annual ACM international workshop on Web information and data management. New York, NY, USA, 56-63.
Tilzer, L. L. and Jones, R. W. (1988). Use of Bar Code Labels on Collections Tubes for Specimen Management in the Clinical Laboratory. Arch Pathol Lab Med, 112, 1200-1202.
Hall, P. A. V. and Dowling, G. F. (1980). Approximate string matching. ACM Computing Surveys, 12(4), 381–402.
Laudon, K.C. (1986). Data Quality and Due Process in Large Interorganizational Record Systems. Communications of the ACM, 29(1), 4-11.

延伸閱讀