IBM’S Watson Wants Your Health Data
The cognitive supercomputer and drugstore chain CVS are partnering on a preventive health platform to reason through 70 million health records.
“While the partnership is just beginning, the ultimate goal is to identify subtle signals of disease progression that allows us to intervene in a more timely fashion and prevent poor outcomes, improving the patient’s health and at the same time reducing healthcare costs,” a CVS spokesperson told The Daily Beast.
According to Fortune, “IBM and CVS are going to work together to create Watson-powered solutions that will then be monetized.”
The partnership aims to transform health care by providing pharmacies, nurses, and physicians with quick responses and best practices for treating patients who endure chronic illnesses such as obesity, heart disease, and hypertension. These conditions are the leading cause of death and disability in the U.S. and account for 86 percent of the nation’s $2.9 trillion in annual health spending.
CVS has over 1,000 walk-in medical clinics, 7,600 retail stores, and some 70 million members with assorted medical plans, making Watson the perfect candidate to deliver new insights and unlock hidden patterns in a massive web of patient-data, CVS says.
Watson’s cognitive computing differs from conventional linear methods because it’s able to make sense of dynamic systems, similar to the way humans do. What makes Watson remarkable, according to IBM, is its ability to stay current by sifting through and interpreting millions of sources, from medical journals to Twitter, within seconds.
Those who are wary of machines learning everything there is to know about a patient’s health ought to know that Watson is heavily reliant on humans to interpret data. Here is how Watson’s cognitive system will interact across various fields in health:
First, Watson builds what’s called a “corpus,” essentially compiling a gigantic literature search on any given subject of interest. Next, with the help of humans, the corpus will be “curated” so only the most up to date and relevant information is stored. Watson then “ingests” or organizes the data by building indices of metadata for fast and efficient retrieval. The last step requires special human attention, where Watson is trained by human experts to interpret information, allowing it to find patterns using a technique called machine learning.
Additionally, CVS will make use of the IBM-operated Watson Health Cloud. It’s a secure, HIPAA-compliant platform for researchers, physicians, and insurers to access individualized patient data. The Watson Health Cloud de-identifies patient information so the privacy and security of individuals remains protected.
“The capabilities of the IBM Watson Health Cloud, when coupled with CVS Health’s insights into medication utilization and patient behavior, could prove transformative for the industry,” said Mike Rhodin, senior vice president at IBM Watson, in a press release.
“Together, we are moving aggressively to design, develop and introduce this new approach to care management—we anticipate that the first version of the solution will be available in early 2016,” a CVS representative said.
Watson is expected to offer health-care providers a holistic view of each and every patient, and its concentration on preventive efforts hopes to save millions in future costs.