In the six years I’ve been working at Penn, this is my first project that’s publicly accessible: Penn Medicine Clinical Trials. It’s a web application to help researchers as well as the general public find clinical trials at the U Penn School of Medicine and the Children’s Hospital of Philadelphia. My co-worker Joe did a great job on the design, and I implemented the functionality. Joe gets the credit for the Google-like “less is more” user interface, which is dramatically different from the clinical trials sites at other leading medical research institutions.
I wrote a script that runs nightly to pull down XML documents on the clinical trials from ClinicalTrials.gov (a website created by the National Institutes of Health to provide a central clearinghouse for information relating to clinical trials in the US). The script then parses out the most important data and puts it in a database, where its used for full text searching. For displaying the pages on each trial, I created an XSL template that’s used to dynamically transform the XML documents for display as web pages.
This was actually part of a larger project – the redesign and relaunch of PennMed’s ITMAT (Institute for Translational Medicine and Therapeutics) web site. Joe created that design as well, and I provided the functionality for the member pages. These pages are dynamically updated as members update their profiles and new members join.
I have a few posts in mind based on this work, as I solved some problems on my own where there weren’t adequate explanations online. The topics are XML handling with PHP’s SimpleXMLItetator, XSL transformations with PHP’s XSLTProcessor, and an interesting CSS problem with the Clinical Trials “sticky footer” that we had to figure out. Stay tuned.