Advances in artificial intelligence (AI) are changing the way people interact with technologies across all kinds of industries. When you navigate using the Google Maps app, receive movie recommendations from Netflix, talk to Siri or use Facebook's facial recognition software to tag friends in a photo, you're using AI.
It's only recently, however, that artificial intelligence has started to make inroads in healthcare. Thanks to breakthroughs in AI, advanced algorithms can now understand and use speech, identify subtle visual patterns and even learn complex skills.
Here are 3 emerging examples of how artificial intelligence could have a major impact on consumer healthcare in the near future. What they all share is the potential to solve a common problem consumers report: what to do when they face certain everyday health issues.
This cylindrical wireless speaker made its US-wide debut in 2015 . What's interesting here is Echo's built-in voice-enabled personal assistant, called Alexa. You can ask Alexa surprisingly sophisticated questions and receive responses in just a fraction of the time it takes Siri to talk to you. She can give you a weather report, play music you want to hear, tell you how many calories there are in a piece of apple pie, read you Wikipedia entries, or even tell a joke. What's especially unique is that Amazon has made the smart move of opening the Echo platform to outside developers, so they can create new "skills" for Alexa. Here's where this artificial intelligence technology is ripe for healthcare innovators.
Boston Children's Hospital was one of the first to take advantage of this opportunity, adding a new skill to Alexa that allows her to give parents simple health advice about a child's fever, medication dosing and when to call the doctor for various common symptoms. Called Kids MD, this Alexa app is the beginning of a plan Boston Children's Hospital has announced for bringing routine, reliable medical guidance quickly to consumers through voice as opposed to print or Web content.
Fitbit, Inc. also recently announced new Alexa-enabled wearable fitness devices that let you ask Alexa how you're doing with your routine and get feedback and motivational tips. The possibilities of virtual health guides like these are exciting.
In 2011, an artificial intelligence computer program wowed the tech world and game show lovers alike by beating two of Jeopardy's greatest champs. That was IBM Watson, which ingested and processed reams of unstructured data and applied it to correctly answer the most tricky, nuanced Jeopardy questions.
More recently, Watson's cognitive computing capabilities are being used in a number of different health applications. One of the first efforts, announced in May 2015, involved uploading huge volumes of clinical data related to cancer to identify evidence-based treatment options based on a patient's medical history, tumour type and other data. Watson has already helped lower the error rate in cancer diagnoses by physicians.
On the consumer healthcare front, Watson is the brains behind some new health-related apps: the Nutrino app powered by Watson gives pregnant women access to personalised guidance on nutrition and food choices, evolving as it learns from them; the SleepHealth from Apple’s ResearchKit, which helps people contribute to and learn how their sleep habits may affect their wellbeing; and UARecord, which helps “coach” athletes on sleep, fitness, activity and nutrition, based on comparative data and outcomes learned from others like them.
This San Francisco-based start-up was named one of the world's 50 smartest companies by MIT last year, thanks to an artificial intelligence system it developed that offers unprecedented pattern recognition skills.
Last year, Enlitic's deep learning algorithm proved better than human radiologists at detecting tiny lung tumours on chest X-rays, increasing accuracy by 50% to 70% and at a speed 50,000 times faster. Same goes for bone fractures. Enlitic's algorithm detected fractures as small as 0.01% of the total bone X-ray image—something no human eye can do. Not surprisingly, hospital systems around the globe are scrambling to work with Enlitic and other artificial intelligence companies in the hopes that these technologies can help diagnose disease faster, cheaper and more accurately than humans.
The holy grail of artificial intelligence—software that mimics human intelligence—remains far off. Artificial intelligence in healthcare will not be able to replace a doctor—at least not anytime soon—but it's becoming clear that these technologies will soon offer personalised healthcare guidance to consumers and enable them to manage more and more aspects of their health on their own.