Development and implementation of a clinical data mining system for gastric cancer surgery / 中华胃肠外科杂志

ABSTRACT

<p><b>OBJECTIVE</b>To develop a clinical database system of gastric cancer surgery integrated with data mining function for better management of clinical data and better performance of both retrospective and prospective studies.</p><p><b>METHODS</b>Core fields for clinical data were determined based on the JGCA(13th and 14th edition) and UICC gastric cancer staging system. Microsoft Visual Basic and VistaDB were used for programming. The database structure was designed according to data mining theory and clinical workflow.</p><p><b>RESULTS</b>After one year of development and refinement, data of over 600 patients from our hospital were retrospectively entered, and function tests were satisfactory. This system was accepted as the database platform for the Chinese Laparoscopic Gastrointestinal Surgery Study Group (CLASS) and was successfully used in the first stage of the Multicenter Retrospective Study of the Feasibility of Laparoscopy for Gastric Cancer among 30 hospitals from both Mainland China and Hong Kong. The data mining function met the requirements, which could carry out complex search with visualized presentation. Descriptive analyses could be performed with the analysis function. Efficient communication among institutions could be executed by data import and export with excellent compatibility and without errors.</p><p><b>CONCLUSION</b>The system has established a clinical database of approximately 4000 fields with data mining function. This system can be widely applied for the clinical research for gastric cancer.</p>