In more than 99% of cases
Watson's recommended treatment plans were the same as those of the oncologists participating in a study.
Why healthcare providers should embrace Artificial Intelligence
September 13, 2017
Contributor
Laura Mueller
Essential Insights contributor, healthcare writer
Featured expert
Jacob Haning
Senior Analyst and Software Engineer, Fuse by Cardinal Health
At Boston Medical Center, a patient advocate named Louise is helping patients with their "After Hospital Care Plans." Only, Louise isn’t your typical hospital staffer. She’s actually virtual.
Louise assists the medical center’s discharge advocates in teaching patients about components of their care, and her dialogue is tailored for each patient based on the information entered into the workstation.
Artificial intelligence (AI) technologies like Louise are becoming a more common part of the patient experience as AI – the simulation of intelligent behavior in computers – grows in healthcare. There were 106 AI startups in healthcare as of February 2017, according to a report from CB Insights. About two-thirds of them launched just last year.
So what does that mean for healthcare providers and their patients?
While there is still debate around what role AI should play in healthcare, there are many reasons healthcare providers should embrace it. Here are three reasons why.
1. Better, quicker diagnostic and treatment decision support
"The same benefits computers provide, like unlimited memory and immediate recall, are what make AI so powerful," said Jacob Haning, a senior analyst and software engineer at Fuse by Cardinal Health, the company's innovation center. “By applying deep learning algorithms to an unlimited amount of knowledge at ever-increasing speeds, we can make highly accurate, informed decisions to produce better outcomes."
As computing power increases and data storage becomes cheaper, computers are able to provide near unlimited capacities for memory and data. AI algorithms make use of that to quickly disseminate and analyze endless volumes of data for better patient outcomes. These are crucial skills in the world of medicine, where new information on how we understand and treat diseases appears in research papers nearly every day.
Even without the heavy workload that today's healthcare providers face, it's impossible for one physician to read and absorb every piece of up-to-date research in their field. But AI can, and as such, is an indispensable tool for doctors to provide patients with the best and most recent information available for better care.
Consider IBM's Watson, an AI tool that is proving particularly useful in the healthcare environment. One iteration, Watson for Oncology, hosts an extraordinary amount of curated oncological data, including more than 290 medical journals, 200 textbooks, and 12 million pages of text.
Recently, Watson was tasked with assisting in treatment support of over 1,000 cancer patients at the University of North Carolina School of Medicine. In more than 99 percent of cases, Watson's recommended treatment plans were the same as those of the oncologists. In 30 percent of cases, Watson provided additional treatment options based on the latest cancer research.
"Tools like Watson support healthcare teams in their decisions," Haning said. "They can get to diagnoses and treatment options faster and help patients in a better way."
2. Greater ability to focus on value-based care
Get a free machine learning primer.
And, sign up for the newsletter that will help your organization thrive in the challenging, dynamic world of healthcare. Industry-wide shifts toward value-based care mean that physicians and other healthcare providers are expected to be more efficient, effective and attentive than ever before. These are heavy expectations considering most providers' workloads and patient volume. AI can offer assistance here too.
Google subsidiary DeepMind is able to take over repetitive, routine tasks that typically take up a large portion of a provider’s time. At London's Moorfields Eye Hospital, DeepMind has been uploaded with a million eye scans, as well as identifying information about the conditions each represents. In the future, they hope this will reduce the time required for experts to go through all aspects of the eye scans by spotting patterns and categorizing results for faster diagnosis, leaving staff with more time to spend with patients.
"Computer programs in general are ideally suited to handle repetitive tasks because of their speed and accuracy," Haning said. "Now, if you can apply context and interpretation, which is the power of machine learning, a part of AI, the list of tasks a computer can optimize becomes much, much longer."
Providers will be able to focus on the more nuanced facets of patient care, like interpreting diagnoses and explaining them in a more meaningful way, which can lead to better patient involvement and outcomes.
3. The human element of healthcare will always be needed
What is AI? (HubSpot)
While it's certainly true that AI could replace longstanding healthcare tasks like medical scans and data entry, it could also open up a need for even more jobs that require a human touch, like empathizing with patients and guiding them through difficult diagnoses and treatments.
Fear of AI, Haning added, comes from a misunderstanding of the facts. Many people take Elon Musk's argument that we need to be wary of incorporating AI into large industries to mean that AI will eventually operate outside of our control. But what Musk is saying, Haning explained, is that AI is going to be a disruption, so it's important that we incorporate it in a way that's thoughtful and sustainable.
Haning added that the incredibly positive potential for applications of AI in healthcare should also be noted in the global discussion.
Experts also see AI as complementary to doctors’ roles and a way to fill in gaps in care. Venkat Rajan, global director of Frost & Sullivan's Visionary Healthcare Program, told Fast Company that robots won't take over doctors' jobs but rather "democratize" diagnosis and care, especially in facilities that have, for example, a general cardiologist rather than a team with different subspecialties.
"In my opinion, the healthcare industry should embrace AI," Haning said. "Our industry should play a major part in the way this technology is adopted and developed, and our voices need to be heard."
More discussions about the adoption and development of AI in healthcare are sure to continue. But for now, it’s clear that AI offers great potential in healthcare and can help providers make faster, better diagnoses and free up more time for quality patient interactions.