What’s holding back machine learning in healthcare?

Healthcare is full of complex data stored in multiple places and changing every day. This makes it an ideal target for the form of artificial intelligence known as machine learning.

Oxford defines machine learning as “the use and development of computer systems capable of learning and adapting without following explicit instructions, using algorithms and statistical models to analyze and draw conclusions from data patterns.

In recent years, machine learning has already proven useful in diagnosis and can contribute to the efficiency of medical coding. But there are plenty of other places where machine learning can be useful but hasn’t made headway yet. Why is that?

Harshith Ramesh is co-CEO of Episource, a provider of risk adjustment services and software for medical groups and health plans, and an expert in machine learning. We interviewed him to discuss why machine learning is so suitable for healthcare, how it has helped with diagnosis and coding so far, and most importantly, what’s holding it back in healthcare. health.

Q. You argue that healthcare is uniquely suited to machine learning. Why?

A. Machine learning is a branch of artificial intelligence that uses data to mimic the way humans learn, continuously improving the performance of a given task over time. In the healthcare industry, this technology is used to detect patterns in patient health information and refine its algorithms to become more accurate as it learns from the available data.

As more and more provider organizations take on risk under value-based contractual models over the next few years, it has become more important than ever to effectively measure patient outcomes, accurate and cost effective. Machine learning is a key tool that providers can leverage to achieve this goal.

Healthcare is uniquely primed for machine learning due to the exponential increase in the volume of patient data over the past two decades. Today, approximately 30% of global data is generated by the healthcare sector.

This is due, in part, to the widespread use of the electronic health record, which began to gain traction in the 1990s. The digitization of patient information has not only increased the amount of existing data, but has also made easily accessible for machine learning applications.

Beyond EHRs, health data is also generated by an increasing number of sources, such as medical devices, wearables, data clearinghouses, laboratories, and vendor offices. This rich abundance of data is essential for machine learning models to become more accurate in predicting patient outcomes. This can help provider organizations develop a more complete picture of a patient’s health over time.

Health data is also more objective in nature than data generated by other industries, making it particularly compatible with machine learning technology. This is due to standardized procedures, automated systems, medical coders and expert doctors – all of which help eliminate subjectivity from data as much as possible.

For example, industry has established standardized data sets for healthcare organizations to use, such as International Classification of Diseases (ICD-10) codes for diagnostic information or National Drug Codes (NDCs) for drug identification.

Regulations on how healthcare organizations can house and transport data in EHRs have also made it easier for machine learning-powered models to analyze data, discover patterns, and apply algorithms. to improve patient outcomes.

Q. How can machine learning help healthcare provider organizations with diagnosis?

A. Machine learning has a variety of applications in the clinical space. One such application is predictive modeling, which is a commonly used statistical technique that can be used to predict future behavior.

Using predictive modeling, providers can effectively predict whether or not a high-risk patient might develop sepsis or another type of complication after a procedure. This can help determine if they want to take additional preventive measures to mitigate this risk, such as calling patients for regular checkups or optimizing resources to target potential high-risk patients.

It can also support population health management by creating dynamic cohorts, which segment member populations based on a given set of health conditions or other type of pattern. These learnings can then be shared with care management teams, who then determine which interventions would have the most impact for a given cohort.

Finally, machine learning models can help providers perform clinical suspicions. This technology can be leveraged to analyze diagnostic data to predict which patients are most in need of emergency care and identify gaps in their medical history.

Machine learning can also help providers determine if a particular treatment would be effective for a patient, for example by analyzing a patient’s entire medical history to find the safest and most effective drug that a doctor can prescribe based on the diagnosis.

Q. How can machine learning help healthcare provider organizations with medical coding?

A. Providers are often thorough in their documentation processes, but translating this data into just one of the more than 72,000 ICD-10 diagnosis codes can be difficult.

As provider organizations strive to improve data quality, they may choose to use and scale AI technology to help improve the efficiency and quality of medical coding throughout. along the risk adjustment continuum – prospectively, simultaneously and retrospectively.

Before and during the visit, machine learning algorithms can quickly analyze the patient’s medical information and present the provider with a real-time snapshot of the patient’s health status.

Clinicians can spend less time on tedious administrative tasks and instead spend more time providing focused and timely patient care. Additionally, prospective coding powered by machine learning may surface chronic conditions that have been documented in the past but not at the time of the visit.

Machine learning can intelligently and automatically analyze unstructured information in the EHR to identify the most accurate code. For example, it can also be used retrospectively to increase both the speed and accuracy of coding, saving provider organizations time and money, allowing them to direct more resources to where they are. most needed.

This, in turn, helps provider groups meet quality metrics, track performance, and ensure patients are assessed regularly.

Q. What is stopping healthcare from making more progress with machine learning?

A. The single biggest factor contributing to healthcare’s reluctance to adopt machine learning is the barriers the industry faces in becoming more interoperable. The competition and resulting lack of coordination between health systems has resulted in myriad challenges.

Whether it’s inconsistent technical standards, differing health information privacy policies, differing approaches to obtaining patient consent, or difficulty coordinating key EHRs, healthcare organizations must overcome many hurdles in their quest for interoperability.

This creates a data gap between different EHR applications and networks, creating silos in data that would inform the most urgent and impactful interventions for patients.

To compound all of this, the healthcare regulatory landscape continues to grow increasingly complex, with revisions to the rules for government-sponsored programs on an annual basis. This adds to supplier doubts about whether technologies such as machine learning can adapt to these constant regulatory changes.

Skepticism will always exist of any emerging technology, especially when providers are presented with one-size-fits-all black box solutions that don’t effectively equip them to deliver better care to their patients.

Technology vendors offering solutions that leverage machine learning technology must transparently explain how they can improve workflow efficiency and reduce administrative burden, giving vendors more time to focus on providing care.

Suppliers should serve as a constant resource and partner throughout the implementation process and beyond, ensuring that their solutions are continuously working to better understand the member population of the supplier organization and improve results for the patients.

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

Sherry J. Basler