While artificial intelligence in healthcare holds great potential, currently though, only twenty-five percent of providers are using artificial intelligence in their data processing. Although there are some drawbacks, like the expense for example to integrate artificial intelligence into patient profiles, Steve Whitehurst, CEO of Health Fidelity, predicts that the entire market will be using AI to analyze data within 2-3 years, because the return on investment is so high.
Health Fidelity provide scalable risk adjustment solutions and Digital Journal caught up with Steve Whitehurst to discover more about artificial intelligence in healthcare.
DJ: How common is artificial intelligence becoming in healthcare?
Steve Whitehurst: The buzz around AI and machine learning platforms is steadily rising. Many applications promise to revolutionize health care in terms of improving patient care with quicker, more precise diagnoses and treatment, and in streamlining the business processes that happen behind the scenes – and they’ve shown great potential in academic research, as well as limited clinical settings so the excitement is valid.
However, in healthcare the stakes are impossibly high – there is literally no room for error – so the technology still has a long road ahead until it emerges as real-world, productized solution for augmenting real-time clinical decision making.
Most of the ready-to-use application thus far have been toward modernizing the underpinning processes of healthcare. Our company, Health Fidelity, is currently leveraging Natural Language Processing (NLP) to analyze patient data for creating a more accurate and complete risk profile of patients. NLP is a subset of artificial intelligence; it is a technology that is able to understand human language.
DJ: What are the advantages?
Whitehurst: AI should be treated as one of the many tools at the disposal of the user, not the definitive solution in and of itself. AI is a tool that enhances our capability, allowing humans to do more than what we could on our own, offering heretofore unattainable speed, scale, and accuracy. However, it’s designed to augment human insight, not replace it.
For example, a doctor can use AI to access the distilled expertise of hundreds of clinicians for the best possible course of action. This is far more than he or she could ever do by getting a second or a third opinion.
DJ: Are there any drawbacks?
Whitehurst: Black box AI solutions that offer no explanation into why they suggest a specific course of action can be at best untrustworthy and at worst downright dangerous, particularly in healthcare. There can be myriad confounding factors in a patient’s condition, some of which may not be obvious as part of the medical record.
AI tools must be able to provide evidence as to how they arrived at a specific conclusion, which allows providers to verify that the conclusion makes sense and course correct if necessary. This optimizes two somewhat divergent goals: improving trust in the output of AI-based systems while at the same time preventing blind faith in them.
DJ: Are there any particular security concerns?
Whitehurst: Many are concerned about the risks; data security, privacy, and regulatory compliance are only part of the issue—the real risk could be one of patient safety. If an algorithm designed to help improve care instead leads a provider astray, to a wrong conclusion or incorrect diagnosis, it could cost patients their lives.
While the results of algorithms are never applied directly to patients without human review, cognitive load on providers is a very real thing. For this reason, poor guidance from machine learning can cause unanticipated harm.
A machine learning system can generate poor results due to many factors. They could be accidental, such as bad input data, bad processing, a software glitch or a bug in the code. That’s not to say all AI and machine learning solutions are risky—merely that organizations need to understand exactly what they’re buying with regard to capabilities and risk potential.
DJ: What services does Health Fidelity provide?
Whitehurst: When Health Fidelity was formed, we realized that the majority of the data in the electronic health record (EHR) – about 80 percent – was unstructured, meaning it could not be readily utilized digitally. The majority of clinical information about a patient lies in the physician’s narrative notes (what we call unstructured data), and nobody was using it. A central principle in founding the company was to make use of that data, and that brought us to natural language processing.
Fast forward to today, Health Fidelity is focused on helping health plan and provider organizations address risk and quality gaps in value-based payment models through NLP-powered software applications. The NLP engine digests, analyzes, organizes, contextualizes, and prioritizes medical record data in its entirety, paving the way for advanced analytics and insights, which in turn help our clients improve their clinical and financial outcomes. In addition to our technology solutions, we also provide advisory services to help clients develop a customized plan for modernizing and transforming their infrastructure and operations needed for value-based care.
DJ: Who do you work with?
Whitehurst: Health Fidelity has over 30 direct client relationships with both health plans and providers of varying sizes and types including some very prestigious organizations like the University of Pittsburgh Medical Center, and Columbia Technology Ventures, a division of Columbia University, who is our NLP partner.
DJ: What other projects are you working on?
Whitehurst:One of the biggest challenge we face in our space is getting access to the EHR data so that our NLP engine can extract insights and provide value to our clients. Much of this is currently done manually – via fax, scan, email, mail, or even sending a record abstractor physically to a doctor’s office – and we believe there’s an opportunity to disrupt this. This is a pain point acutely felt by all of our clients and partners, so we’re working on developing ways for technology to enable better data sharing needed to succeed in the value-based care world.