Unlocking Data for Clinical Research - The German i2b2 Experience

被引：39
作者：
Ganslandt, T. [1 ]
Mate, S. [2 ]
Helbing, K. [3 ]
Sax, U. [3 ,4 ]
Prokosch, H. U. [1 ,2 ]
机构：
[1] Erlangen Univ Hosp, Ctr Med Informat & Commun, DE-91054 Erlangen, Germany
[2] Univ Erlangen Nurnberg, Chair Med Informat, D-91054 Erlangen, Germany
[3] Univ Med Ctr Gottingen, Dept Med Informat, Gottingen, Germany
[4] Univ Med Ctr Gottingen, Div Informat Technol, Gottingen, Germany
来源：
APPLIED CLINICAL INFORMATICS | 2011年 / 2卷 / 01期
关键词：
Medical Record Systems; Information Storage and Retrieval; Single-Source;
D O I：
10.4338/ACI-2010-09-CR-0051
中图分类号：
R-058 [];
学科分类号：
摘要：
Objective: Data from clinical care is increasingly being used for research purposes. The i2b2 platform has been introduced in some US research communities as a tool for data integration and querying by clinical users. The purpose of this project was to assess the applicability of i2b2 in Germany regarding use cases, functionality and integration with privacy enhancing tools. Methods: A set of four research usage scenarios was chosen, including the transformation and import of ontology and fact data from existing clinical data collections into i2b2 v1.4 instances. Query performance was measured in comparison to native SQL queries. A setup and administration tool for i2b2 was developed. An extraction tool for CDISC ODM data was programmed. Interfaces for the TMF privacy enhancing tools (PID Generator, Pseudonymization Service) were implemented. Results: Data could be imported in all tested scenarios from various source systems, including the generation of i2b2 ontology definitions. The integration of TMF privacy enhancing tools was possible without modification of the platform. Limitations were found regarding query performance in comparison to native SQL and certain temporal queries. Conclusions: i2b2 is a viable platform for data query tasks in use cases typical for networked medical research in Germany. The integration of privacy enhancing tools facilitates the use of i2b2 within established data protection concepts. Entry barriers should be lowered by providing tools for simplified setup and import of medical standard formats like CDISC ODM.
引用
收藏
页码：116 / 127
页数：12
相关论文
共 29 条
  • [1] [Anonymous], 2013, The Data Warehouse Toolkit
  • [2] CDISC, 2010, OP DAT MOD ODM
  • [3] Description of a Rule-based System for the i2b2 Challenge in Natural Language Processing for Clinical Data
    Childs, Lois C.
    Enelow, Robert
    Simonsen, Lone
    Heintzelman, Norris H.
    Kowalski, Kimberly M.
    Taylor, Robert J.
    [J]. JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION, 2009, 16 (04) : 571 - 575
  • [4] Evaluating the informatics for integrating biology and the bedside system for clinical research
    Deshmukh, Vikrant G.
    Meystre, Stephane M.
    Mitchell, Joyce A.
    [J]. BMC MEDICAL RESEARCH METHODOLOGY, 2009, 9
  • [5] DIMDI, 2010, INT CLASS DIS ICD10
  • [6] An optimal code for patient identifiers
    Faldum, A
    Pommerening, K
    [J]. COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE, 2005, 79 (01) : 81 - 88
  • [7] Gainer Vivian, 2007, AMIA Annu Symp Proc, P959
  • [8] Goryachev Sergey, 2006, AMIA Annu Symp Proc, P931
  • [9] Medical i2b2 NLP smoking challenge: The A-life system architecture and methodology
    Heinze, Daniel T.
    Morsch, Mark L.
    Potter, Brian C.
    Sheffer, Ronald E., Jr.
    [J]. JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION, 2008, 15 (01) : 40 - 43
  • [10] Helbing K, 2010, METHODS INF MED, V49
123