Users can use voice to describe conditions and connect with the proper information or personnel for diagnosis and treatment.
Expect Labs, provider of the MindMeld platform for building artificial intelligence-powered, voice-driven applications, has partnered with Sense.ly, providers of a virtual nurse application that provides personalized patient monitoring and follow-up care, to build customized virtual assistant applications for the healthcare industry.
The first product to result from that partnership is a mobile virtual nurse that is voice-enabled. The virtual medical assistant, named Molly, can automate many of the most common patient services normally delivered by doctors. Patients in hospitals and clinics can use Molly for long-term maintenance of chronic conditions. Molly can assess them for risks, offer education and insights, and tell them if they need to see a doctor. If necessary, the application can even connect the patient to a real human nurse or doctor via a video chat session.
Patients interact with Molly using simple voice commands and can ask a wide range of health-related questions. MindMeld's AI technology is used to understand the meaning of the spoken questions and provide answers or relevant medical information.
With this smartphone solution, a patient using Sense.ly will be able to ask questions such as, "I have ringing in my ears and am feeling dizzy and nauseous. What should I do?" MindMeld will then understand the meaning of the question and determine what the patient might be experiencing. Through advanced language understanding technology, MindMeld can provide detailed information about possible conditions and where to turn to for help.
The application searhes a number of sources, including the Web, government databases, and patient records, for answers to the questions posed by patients, according to Tim Tuttle, CEO of Expect Labs.
The application currently supports tablets and smartphones running on Apple's iOS mobile operating system. Healthcare networks buy the technology and provide it to their patients for free.
"Healthcare providers want to offer an app like this because over time it reduces the long-term costs associated with caring for patients with chronic conditions," Tuttle says.
In early trials of the technology, engagement has been very high, with as many as 80 percent of the patients who have used Molly completing all conversations without dropping off. In places where it's been used, the app has reportedly reduced patient time by about 20 percent.
"We created Sense.ly to put an intelligent virtual nurse in the palm of every patient's hand to make healthcare easy and accessible," said Adam Odessky, CEO of Sense.ly, in a statement. "By complementing physician expertise with advanced machine learning, MindMeld is helping us enable healthcare providers to deliver better care at lower costs."
The healthcare vertical is an important proving ground for the technology, Tuttle points out. Advances in machine learning have expanded the types of questions and the language that applications of this nature can handle, he says, but in many cases, patient needs also play a role.
"Many users in [the healthcare] domain might have disabilities that make voice an essential interface," Tuttle explains. "Most are not computer-savvy but still need access to information, so voice becomes a very important feature."
In fact, speech "is one of the most liked and widely used features," he continues. "Patients find using voice much more convenient than typing."
That transcends the healthcare industry, according to Tuttle, who notes that Google now claims that 10 percent of all search queries are coming through voice, and Apple recently announced that Siri fields more than 1 billion inquires a week. "Consumer adoption and acceptance of voice has skyrocketed," he says. "It's a huge trend right now."