2016 saw significant improvement in the Text Information Extraction System (TIES) software. We released two new versions (5.5 and 5.6), and also welcomed Thomas Jefferson University and Stony Brook University into the TCRN network. Additionally, TIES and the TCRN members published several papers on cancer research and Noble Coder (a natural language processing component). TIES also gained new social media updates, website page updates, and blog entries to give users a look behind the TIES code. All of this this progress helped add momentum to the many TCRN nodes collaborations, supporting data and tissue exchange across the network.
Our March 23rd release of version 5.5 included new LDAP support, pop out reports in separate windows, numbering patient codes, and regulatory administrator audit features helped TIES become a faster and more competent service. Multiple bug fixes included diagram mode overhauls to support better branch organization and revamped term highlighting for queries unable to highlight certain entries increased user readability and access to tissue samples.
In September, we added manual annotation, making it possible for users to add their own annotations to cases and case sets and share information between other colleagues working on the same projects. Temporal queries also saw advancement in identifying which case set belonged to which patient, making multiple cancer searches or follow-up documentation for patients easier to locate. Finally, TIES users are now able to export Excel files for easier annotation and case set comparison. Together with numerous other bug fixes, TIES upgraded to a new and more advanced version of itself.
Our 2016 efforts also included a reinvigorated social media campaign, which included more access to the inside of what TIES developers were working on, what the newest software features were, and what TIES hoped to produce in the future. Blog entries and a regularly updated Twitter feed allowed TIES to get closer to its user base.
Published and Accepted Papers
TIES started 2016 off well with a published paper by Eugene Tseytlin, et al. titled NOBLE, a flexible concept recognition for large-scale biomedical natural language processing (access paper here). In the later part of the year, Thaer Khoury, MD, FCAP, a TIES user at Roswell Park informed us that his paper on the role of ulceration in breast cancer became our first scientific paper to include data collected across the entire network. Several other pilot projects are ongoing. And we expect an increasing number of publications from TCRN users as our network grows.
Thomas Jefferson University and Stony Brook University
Finally, on May 25, 2016 Thomas Jefferson University became an official member of the TCRN. They planned to archive nearly 950,000 reports from 2004 to the present. These reports include surgical, cytology, biopsies, molecular reports, and autopsies. Additionally, Stony Brook University completed their Network Agreement and became a full-fledged TCRN member. Once they finalize their deployment, Stony Brook will become the 6th TCRN site.
While TIES has grown by leaps and bounds in 2016, there is still a long way to go to achieve our vision of a nationwide data, image and biospecimen network to support cancer research. Stay tuned for a look into what 2017 will bring to TIES.