Pathology Tumor Reporting & Data Depository using a Web-based System
Zhenhong Qu MA; University of Rochester; Kathryn E. Schafer MS; University of Rochester; Jennifer Findes-Hosey MA; University of Rochester;
Content:
Context: The American College of Surgeons Commission on Cancer (CoC) mandates that pathology reports at CoC-approved cancer programs include all scientifically validated or regularly used data elements for each site and specimen. To increase user compliance, we have created a web-based tumor reporting system that can facilitate tumor reporting in pathology. The re-usability of the collected data, however, is limited by en block text data entry into lab information system (LIS). We now modify the existing system in an attempt to enhance the usability of the collected data by the end users on the World Wide Web.
Technology:
Technology: We use Microsoft (Microsoft Corporation, Redmond, WA) Access to generate tumor reporting templates, and create an emulating executable database program that can be downloaded by Internet users. Microsoft Active Server Page (ASP.NET) is used to present the template as dynamic webpage on the Internet, and compile the selected information into a text file. It is also used to code a middleware that can import text files into the downloaded database.
Design:
Design: We modify the underlying architecture of our existing Web-based tumor reporting system (Qu et al., AJCP 127:898, 2007) by 1) incorporating SNOMED-CT and ICD9, and 2) adding a downloadable database emulating the on-line reporting system and a middleware which can import the reporting data elements into the database by the web user.
Results:
Results: Internet users can generate a tumor report by filling out each template with data elements in dropdown lists. After the report is compiled, the data elements can be incorporated into pathology report in local LIS by copy-paste. In addition, the collected data for each case can be saved locally as delimited text file. The emulating database and the text-parsing middleware can be downloaded. The middleware can import the delimited text files (= tumor reports) into the database independent from the local LIS or the web-based reporting system. The locally collated report data can then be mined or imported into the users LIS.
Conclusion:
Conclusion: The new system 1) can facilitate routine tumor reporting, 2) enhance data retrieval by SNOMED / ICD9 codes from local LIS, and 3) allow local structured data depository by the end users. The final results would be increased reporting accuracy and enhanced data re-usability.
