Although healthcare is squarely in the era of big data and data analytics, it remains difficult in clinical research to accurately identify patients with complex conditions like valvular heart disease through medical records.

THE PROBLEM

And if researchers cannot identify these patients, they cannot study them, track practice patterns or conduct population management.

Part of the problem is that the current methods used to identify highly specific conditions like valvular heart disease use diagnosis or procedure codes. These were created primarily for billing purposes and often are not very useful for clinical care because they can be quite nonspecific and not include detailed data about the condition.

"For example, a patient with moderate or severe aortic stenosis, which is a narrowing of one of the primary heart valves, is entirely different than a patient with mild valve disease," said Dr. Matthew Solomon, a cardiologist at the Permanente Medical Group and a physician researcher at the Kaiser Permanente Division of Research in Oakland, California.

"Yet some of the codes simply use 'aortic valve disease,' which could be applied to an entirely different clinical problem," he continued. "Without accurate and systematic case identification, population management and research for valvular heart conditions isn't possible."

In addition, the data needed to identify patients with valvular heart disease is buried in echocardiography reports, which, like many radiology reports, often are free-text fields that are heterogeneous and unstructured, and cannot be easily queried.

"Now we are using this data to examine our practice patterns and outcomes for these patients so that we can improve our care and understanding of these patients for ourselves and the broader medical community."

Dr. Matthew Solomon, the Permanente Medical Group

"The only solution to identify these patients was either to have an army of humans pore over 1,000,000 echocardiography reports, or by developing natural language processing methods and to teach a computer how to do that for us," he stated.

PROPOSAL

NLP is a branch of AI where a complex set of rules is developed to read unstructured, free-text reports, and to then create a structured, systematic and organized database. Once this is done, the potential for both studying this population and performing high-quality population management becomes possible.

The development and validation of such methods is a growing field, and Kaiser Permanente Northern California Division of Research is at the forefront.

MEETING THE CHALLENGE

"My colleagues and I used a software application as the architecture to build and validate our NLP tools, but the algorithms were not something we found or borrowed off the shelf," Solomon explained. "These algorithms were then applied to our entire dataset within the EHR. This involves organizing the data from our backend EHR systems, and then running the formatted data through the software to create an organized and structured dataset.

"We currently are conducting research on patients with valvular heart disease, and we are moving to incorporate these methods to identify patients, in real time, to establish one of the largest population management programs in the world for this patient population."

RESULTS

The researchers' first achievement was the successful development and validation of the technology, the results of which they published this year in the Journal of Cardiovascular Digital Health.

"Our next step was to apply the technology to our echocardiography database within Kaiser Permanente Northern California, which included nearly one million echocardiography reports from the past decade," Solomon reported. "In minutes, the software identified nearly 54,000 patients with the condition, a process that would likely have taken years for physicians to perform manually.

"Not only did we identify the patients, but we were able to also extract all the key detailed elements from each echocardiography report," he added. "Now we are using this data to examine our practice patterns and outcomes for these patients so that we can improve our care and understanding of these patients for ourselves and the broader medical community."

ADVICE FOR OTHERS

"These AI techniques are extremely powerful, and they are helping us to shepherd in a new era in our ability to use big data and analytics in healthcare to better serve our patients," Solomon concluded. "I would highly encourage all healthcare organizations to invest in the people and technology who can carry out this work. It truly brings healthcare into the 21st century."

Twitter: @SiwickiHealthIT
Email the writer: bsiwicki@himss.org
Healthcare IT News is a HIMSS Media publication.