NCRA’s “Mountains of Change” Part II: Artificial Intelligence | Blog

NCRA’s “Mountains of Change” Part II: Artificial Intelligence

NCRA’s “Mountains of Change” Part II: Artificial Intelligence

Tue, Jul 9, 2019  -  Comments (0)  -   Posted by Karen Mordarski

NCRA’s 2019 Educational Conference in Denver, Colorado, was built around the theme, “Navigating the Mountains of Change.” And during the four days of sessions I attended, three “mountains” stood out to me as particular challenges for CTRs to navigate:

Precision Medicine & Genetics | Artificial Intelligence | Biomarkers

Last month I posted on precision medicine and genetics. Today, I will share my notes from an NCRA session on the second mountain, artificial intelligence, which is another new frontier that can greatly help the cancer registry. Karen Mason, RN, MSc, CTR, and Kelly Merriman, PhD, MPH, CTR, presented “The Cancer Registry in the Era of NLP” which went in-depth on the topic of AI and the cancer registry. Here’s what I took away from their presentation…

WHAT DOES ARTIFICIAL INTELLIGENCE MEAN FOR OUR PROFESSION?

Currently, all dictated records within the EMR are considered unstructured data. The cancer registry is evolving by collecting more relevant cancer data, thus increasing the complexity in abstraction. CTRs need help with managing increased workloads, keeping up with the current dataset and, potentially, responding to more revisions and demands for real-time data. Advances in IT and the dawn of artificial intelligence, specifically natural language processing (NLP), will be integrated into healthcare EMRs and become extremely helpful to CTRs.

SNOMED CLINICAL TERMINOLOGY

In cancer, research is exploding with an emphasis on genomics and translational medicine to extrapolate the data combined with genomic data to discover new clinical pathways for drug protocols and treatments tailored to individualized care within the precision medicine model. The emphasis is on automating cancer registration through NLP of structured data into the registry’s database through recognition of key phenotype variables within the abstract. The most logical place to start is with the pathology report and SNOMED clinical terminology. SNOMED CT is the most comprehensive and precise multilingual health terminology in the world. A healthcare system can use NLP to extract and data mine for cancer casefinding and populate an abstract; then a CTR can choose the correct record and fields, validate, and quality assess the abstract. The primary function of NLP is to extract valuable clinical information embedded in transcribed notes to enhance EMRs, decrease manual effort, decrease error rates and facilitate integration. The NLP pipeline data flow is complex, but if done right, produces “big data” which has contributed to major advancements in medicine in the past.

BETTER DATA SAVES LIVES™

The Department of Energy and an integrated team from NCI’s SEER Program are currently working on a joint collaboration to create NLP and machine learning tools that can accurately capture information from unstructured clinical text for expanding cancer surveillance data reporting. As we know at CHAMPS Oncology, better data saves lives™, and artificial intelligence has the potential to drastically improve the data we collect.

Did you attend this year’s NCRA educational conference? What were your key takeaways? Check back soon for part three of my series on NCRA’s mountains of change.

Interested in learning how CHAMPS Oncology can help your cancer program navigate recent industry changes? Contact us.

Posted in Cancer Registry

Leave a Comment

Your email address will not be published. *Required fields.

     

Tag Cloud