It’s well documented that the healthcare industry is a veritable treasure trove of data but much of this is in unstructured form—for example, clinical notes or patient narratives. Traditionally it’s been difficult for stakeholders to derive meaningful insights from this information, but, according to a new report in the Journal of the American Medical Informatics Association (JAMIA), this is changing thanks in large part to artificial intelligence (AI).
As HealthcareITNews’ Mike Miliard wrote, “The research in JAMIA suggests that real-world data captured in unstructured notes offers more accuracy when trained algorithms are used to mine it.” The study used six years’ worth of de-identified electronic health record (EHR) data to mine a specific set of clinical concepts from both structured and unstructured data, using standard query techniques for the former and AI for the latter.
The HealthcareITNews piece breaks down the study in greater detail, but the researchers ultimately concluded that unstructured data was more useful in predicting coronary artery disease than its structured counterpart. These findings have big implications for the healthcare industry, and other sectors as well.
As AI tools have matured, companies have uncovered numerous new avenues for exploring and leveraging unstructured data, including:
- Natural language processing, which can be used to extract meaning out of written documents like emails, journal entries, news articles, and social media posts.
- Speech-to-text conversion, which can be used to turn audio speech into searchable text.
- Pattern recognition algorithms, which can be used to identify people, animals, or other objects in digital images.
In order for these and other AI-fueled technologies to successfully mine unstructured data for business intelligence, it’s critical that organizations have the right data storage and analytics strategies in place. A traditional approach that keeps data siloed and inhibits widespread sharing and analysis of information will never support the culture necessary for unlocking the value of unstructured data.
To quote the healthcare study’s authors, “The clear learning from this study is that accuracy is heavily influenced by data and technology choices.” Only when equipped with the right technologies can organizations put their best foot forward when analyzing unstructured data.