In a recent webinar, Kata Healthcare CEO Matt Hollingsworth shares how his health tech business uses artificial intelligence for data mining to mobilize medical data so that organizations can transform the patient experience.The company aims to ease the burden on healthcare organizations of bringing structured and unstructured data together in a standardized format so all of their data can be used effectively and consistently across the organization – ultimately improving patient care.
Rachel Ford Hutman, founder of Ford Hutman Media, moderated the discussion.
Hollingsworth highlighted the need for patients with complex medical conditions, such as his mother, to bring their binders of healthcare data to healthcare appointments because Their medical history is not readily available to their doctors. His company seeks to change the status quo in healthcare data availability Automate using natural language processing.
On the other hand, Hollingsworth points out that while natural language processing and automation are important tools healthcare organizations are leveraging, every healthcare facility stores and processes data differently.this means Requires clinical expertise of healthcare clinicians Balancing the Limits of AI in Discrimination Where appropriate information is stored. Likewise, artificial intelligence can complement the human ability to process large amounts of information in a short period of time.
“Because the data is messy and complex, and the same data can exist in multiple places and with different degrees of completion, you don’t necessarily get all the data When humans mine data alone, Because people don’t have unlimited time to read unlimited documents,” Hollingsworth said. “There’s always the chance that something will be missed. “
“On the other hand, a lack of clinical knowledge means you can end up with noisy data. For example, lists of questions are notoriously inaccurate. When they are first captured, they are often accurate, but no one Will continue to set resolution dates for things and work out how long the situation has been around. So you can’t necessarily rely on it. But this information is in things like HMP and progress notes. Again, in order to be able to deal with data that exists in multiple places The truth is, you have to have someone who can teach the system where to find highly reliable sources of information. If you don’t, you end up with very noisy and inaccurate data.
The webinar also provides insights on:
- How AI approaches such as machine learning and natural language processing can standardize healthcare data for clinical registration submissions
- The strengths and limitations of artificial intelligence
- Why a Human-Machine+Computer Data Mining Approach Produces the Best Results
- How to provide clinical registry data in real-time for patient care, quality programs, and other internal programs in hospital systems
To listen to the webinar, please fill out the form below.
photo: ipopba, Getty Images
