Wong Stephen T C, Hoo Kent Soo, Cao Xinhua, Tjandra Donny, Fu J C, Dillon William P
Department of Neurology, Radiology, and Neurosurgery, University of California, San Francisco, CA, USA.
Acad Radiol. 2004 Mar;11(3):345-58. doi: 10.1016/s1076-6332(03)00676-7.
Clinical databases are continually growing and accruing more patient information. One of the challenges for managing this wealth of data is efficient retrieval and analysis of a broad range of image and non-image patient data from diverse data sources. This article describes the design and implementation of a new class of research data warehouse, neuroinformatics database system (NIDS), which will alleviate these problems for clinicians and researchers studying and treating patients with intractable temporal lobe epilepsy. The NIDS is a secured, multi-tier system that enables the user to gather, proofread, analyze, and store data from multiple underlying sources. In addition to data management, the NIDS provides several key functions including image analysis and processing, free text search of patient reports, construction of general queries, and on-line statistical analysis. The establishment of this integrated research database will serve as a foundation for future hypothesis-driven experiments, which could uncover previously unsuspected correlations and perhaps help to identify new and accurate predictors for image diagnosis.
临床数据库在持续增长并积累更多患者信息。管理这些海量数据面临的挑战之一是如何从不同数据源高效检索和分析广泛的图像及非图像患者数据。本文介绍了一种新型研究数据仓库——神经信息学数据库系统（NIDS）的设计与实现，它将为研究和治疗难治性颞叶癫痫患者的临床医生和研究人员解决这些问题。NIDS是一个安全的多层系统，可让用户从多个基础数据源收集、校对、分析和存储数据。除数据管理外，NIDS还提供多项关键功能，包括图像分析与处理、患者报告的自由文本搜索、通用查询构建以及在线统计分析。这个综合研究数据库的建立将为未来基于假设的实验奠定基础，这些实验可能揭示先前未被怀疑的相关性，并可能有助于识别新的、准确的图像诊断预测指标。
