International Business Machines (IBM 0.90%) has invested more than $1 billion in Watson, its cognitive computing system, in an effort to turn it into a major source of revenue. Progress has been slow, partly due to the nature of machine learning systems like Watson, which need to be trained before they become useful.
One area where Watson is making some serious progress, though, is healthcare. In May, IBM announced that it was collaborating with more than a dozen leading cancer institutes with the goal of using Watson to provide personalized treatment options to patients.
Watson isn't new to the medical field. In 2011, Watson had been trained with enough data and information to have the knowledge equivalent to a second-year medical student. As Watson is used at the major cancer institutes, devouring patient data and medical literature, the system will become more accurate, with the goal of providing doctors with insights from the data that ultimately improves patient outcomes. With 1.6 million Americans being diagnosed with cancer each year, the opportunity for IBM to make a real impact, and in turn, transform Watson into a major business, is significant.
Paging Dr. Watson
Genetic sequencing, the process of mapping out a person's DNA, can allow doctors to tailor specialized treatments for each individual cancer patient based on specific mutations. However, this process involves a tremendous amount of data -- more than 100 gigabytes per person, according to IBM -- and it can take clinicians weeks to analyze each mutation, along with the relevant medical literature.
Watson can parse all of this data in a matter of minutes, providing doctors with a list of possible treatment options, along with levels of confidence for those options, which may be more effective than standard procedures. Because Watson is a machine learning system, the more data it takes in, the more accurate the system becomes.
Watson certainly isn't a replacement for doctors, as there's no replacement for experience and one-on-one interactions with patients. Instead, Watson acts like an advisor, freeing doctors from tedious data analysis, and allowing a greater number of patients access to personalized treatment options. This could both improve the average outcome for cancer patients, and potentially lower the cost of treatment.
Why IBM is well suited to the task
The biggest challenge facing Watson in terms of adoption is the fact that it takes time and effort to train the system. Watson is the polar opposite of a plug-and-play solution, and it can take years of training before the system is useful for a specific application.
Watson is named after the first CEO of IBM, Thomas Watson Sr., whose son, Thomas Watson Jr., wrote an excellent autobiography about his life at IBM, both working under his father, and as CEO himself. In his book, Watson Jr. refers to IBM as a service company, even though at the time, IBM's core business was selling and renting mechanical punch-card machines and the earliest computers systems, which were meant largely for accounting and bookkeeping tasks.
What IBM really sold wasn't technology, but a promise to solve customers' problems. In the days before computers, IBM would hold week-long courses for its customers to teach them how to use its punch-card machines. IBM wasn't looking to simply make sales, but to ensure that customers got the most out of its machines.
While IBM may look like an entirely different company today, the core ethos is the same. Watson will succeed, not because it's the best technology -- companies like Microsoft and Amazon offer cloud-based machine learning systems, as well -- but because IBM is committed to its clients.
Earlier this year, IBM announced the formation of IBM Watson Health, a new dedicated business unit meant to accelerate the adoption of Watson in the healthcare field. More than 2,000 IBM consultants, medical practitioners, clinicians, developers, and researchers are dedicated to the task, ensuring that healthcare clients get the most out of Watson.
It's taken some time, but IBM's Watson is starting to find its legs. The potential for Watson to vastly improve the quality of care by providing insights to doctors, sifting through an insurmountable quantity of data and information, is hard to quantify. IBM is making every effort to make Watson a key component in the healthcare field. If it succeeds, Watson Health could become a major part of IBM's business.