2022 - How to Unlock Electronic Health Record Notes Using AI for Real-World Evidence

For real-world data (RWD) to be useful at scale to drive business insights, it often needs to be deemed as ‘quality’ data. Transforming real-world data into a quality data asset often requires it to be organized, cleansed, harmonized and essentially structured in a format that makes it usable. While claims data is predominantly in a structured format, electronic health record (EHR) data contains both structured and unstructured data. In fact, it is estimated that 80 percent of this type of data is unstructured*, often in the form of free-text.
This notes-based data can include important clinical information, such as patient sentiment, genetic testing results, reasons for therapy choice, side effects of a medication, patient outcomes and more. Structuring this free-form data has the immense opportunity to provide deeper insights into many aspects of a patients’ disease journeys, adverse events identification, burden-of-disease tracking, patient identification for a clinical study and more. Artificial intelligence (AI) techniques, such as machine learning (ML) and natural language processing (NLP), paired with oversight from clinicians, are already being used to standardize EHR data in a meaningful way at scale.
Join this webinar to understand how unstructured EHR data is curated using clinician-informed AI to help make it meaningful for real-world research purposes.
Bonus: hear real case study examples across ophthalmology, urology and neurology.
*https://www.ncbi.nlm.nih.gov/pmc/articles/PMC63724...