You might think of artificial intelligence (AI) in health care as science fiction — something that’s a future possibility. But the truth is, the future is here now.
AI is a collection of methods and technologies for performing tasks we often associate with the human mind, such as learning and reasoning. AI approaches already have many applications in health care. They help people with diabetes manage their blood sugar, automate prescription refills and match customers to the person most qualified to assist them by phone, email or text.
Interest in other health care applications continues to grow. About 60 percent of health professionals in a 2017 Forrester study conducted on behalf of Intel said AI adoption was a top or high priority for their organization.
Why is AI so intriguing in health care? As I’ve explored previously, the hype often plays on fears of machines replacing doctors, which is more fake news than reality. Let’s focus for a moment on the AI applications that are practical and possible in the short term.
AI systems can improve operational speed, capacity and consistency and automate repetitive or tedious tasks like processing paperwork. These uses enable health care professionals to practice at the top of their license.
Health information management (HIM) specialists are one group currently realizing value from AI applications. These professionals working in hospitals, health systems and physician practices are highly skilled and responsible for managing vast amounts of data, along with the reimbursement frameworks within care delivery organizations. However, they often spend too much time on lower-order tasks versus activities that leverage their domain expertise when assessing the completeness of clinical documentation and ensuring appropriate code assignment for medical claims. AI can automate much of the work in these reviews and flag cases that have gaps in documentation, which affect coding and reimbursement.
Health information management isn’t the only place AI can improve efficiency and accuracy. It can reduce the administrative burden on physicians, nurses and allied health professionals. This can be significant, considering that one report calculated doctors spend one-sixth of their working hours on administrative work. With speech-based interfaces, AI can automate some of the data entry that burden many physicians, apply clinically aware models that prioritize both clinical and administrative work queues to optimize workflow, and even offer physicians decision support recommendations based on data-driven insights and evidence-based medicine. These AI applications enable clinicians to provide better care and give their patients more attention.
When you are exploring how to apply AI in your business, there are some key items you should address to set your organization up for success. Decide what you will use AI for and what the expectations of it will be. Then, plan and implement your AI integration using these fundamental steps:
- Identify use cases: What do you need or want to improve, and is AI the right solution? It’s important to focus, because AI is not a magic wand that can fix all problems at once.
- Convene the right people: Get data scientists on board, and also think about the end users. Who needs the information your AI application provides, and how will they consume and act upon that information? The technology should bring usable data and recommendations to people involved in areas like clinical work, quality control and revenue cycle.
- Curate data: Predictive accuracy of a system increases from below 1 percent to over 90 percent by including information from a variety of sources, according to a Frost and Sullivan study. But that data needs to be standardized and organized to unlock all the key information AI needs.
- Set benchmarks for success: How do you measure whether your AI implementation is solving the problems you intended it to? Set goals and measure whether you are achieving them, so you can continually improve and adapt to change.
The potential and promise of AI continue to grow as more applications and integrations are undertaken. Explore this interactive graphic for inspiration and details about how it’s already being used and its biggest potentials for providing real value in health care.
Mark Morsch, vice president of technology for Optum, is an innovation leader with a focus on developing products using natural language processing and other forms of artificial intelligence to help transform our health care system. Guided by the goal to deliver the right information at the right time to the right audience. Inspired to build world-class development teams that can drive innovation and surpass client expectations through creativity, engagement and hard work.