The Washington PostDemocracy Dies in Darkness

How IBM Watson will impact our fight against cancer

Leanne LeBlanc, IBM Watson project manager, views analytics of health-care data at Watson headquarters in New York City. IBM launched the new Watson Health business unit to help patients, physicians, researchers and insurers use data to achieve better health and wellness for all. (Jon Simon/Feature Photo Service for IBM)
Placeholder while article actions load

At the inaugural World of Watson event in New York Tuesday, IBM announced a new Watson Genomics initiative that will utilize the computing capabilities of Watson to make it easier and faster to fight cancer. It will now be possible for clinicians at more than a dozen cancer institutes around the nation to apply Watson’s data-crunching abilities to sequence the genome of a cancerous tumor and then access the most relevant information in the medical literature to recommend a treatment option. The goal is an ambitious one: personalized medicine for cancer patients everywhere based on their unique genomic profile.

In the initial phase of the program, participating organizations will apply Watson to analyze the genomic data of patients who are battling all types of cancer, including lymphoma, melanoma, pancreatic, ovarian, brain, lung, breast and colorectal cancer. In cases where surgery, chemotherapy or radiation treatment have failed, these cancer patients could benefit from isolating treatments that target their specific cancer-causing genetic mutations. The collaborations at these institutions — including New York Genome Center, McDonnell Genome Institute at Washington University in St. Louis, University of North Carolina Lineberger Cancer Center and Yale Cancer Center — will enable clinicians to use Watson with an expanded set of patients by the end of 2015.

[The incredible potential and dangers of data mining health records]

The most important factor, say clinicians involved in the new Watson genomics initiative, is the faster processing time of DNA insights made possible by using Watson. In some cases, the processing time can be reduced from weeks to minutes. This is critically important, given the amount of genomic data out there. As IBM pointed out at World of Watson, one person’s genetic code is the equivalent of 1.5 billion characters of text, or 10,714,286 tweets. If you printed out the 3.2 billion letters in your genome, it would take a century to recite, if recited at the rate of one letter per second, 24 hours a day.

“When you are dealing with cancer, it is always a race,” said Lukas Wartman, assistant director of cancer genomics at the McDonnell Genome Institute at Washington University in St. Louis, who spoke at the World of Watson event. “As a cancer patient myself, I know how important genomic information can be. Unfortunately, translating cancer-sequencing results into potential treatment options often takes weeks with a team of experts to study just one patient’s tumor and provide results to guide treatment decisions. Watson appears to help dramatically reduce that timeline.”

Rolling out Watson to more than a dozen cancer institutes around the nation is just the beginning. At some point, any oncologist might have access to Watson, meaning that cancer patients in any geographic locale would benefit, not just those located near a major cancer research institute.

“This collaboration is about giving clinicians the ability to do for a broader population what is currently only available to a small number – deliver personalized, precision cancer treatments,” said Steve Harvey, vice president of IBM Watson Health. “The technology that we’re applying to this challenge brings the power of cognitive computing to bear on one of the most urgent and pressing issues of our time – the fight against cancer – in a way that has never before been possible.”

And best of all, Watson will continue to learn on the job as it hunts for appropriate oncology treatments. That means that Watson will gain in value and knowledge over time, based on previous interactions with medical practitioners. The more that participating institutions use Watson to assist clinicians in identifying cancer-causing mutations, the more that Watson’s rationale and insights will improve.

With the help of Watson, developers and partners are already creating other apps and services that harness the power of cognitive computing to help health care professionals do their jobs better and solve important challenges. At the World of Watson event, for example, Watson Health developers showcased new apps and services built on top of the Watson platform. The medical community has already been one of the earliest adopters of Watson cognitive computing technology, where IBM has collaborated with leading hospitals and research institutes including Memorial Sloan Kettering Cancer Center, University of Texas MD Anderson Cancer Center, the Cleveland Clinic, the Mayo Clinic and the New York Genome Center.

Of course, there are still some challenges ahead for Watson. Processing unstructured data – images of cancerous tumors or notes that oncologists might add to a patient’s medical file – is a lot harder than processing structured data. Physicians shouldn’t have to be programmers and coders to extract insights, something that IBM Watson recognizes. And there’s still the human aspect of working together with patients to agree on the appropriate treatment option.

Keep in mind, however, that the goal of IBM’s cognitive computing initiative for health care is not intended for Watson to replace physicians, nurses and other medical practitioners. The metaphor IBM typically uses for Watson is one of a “capable and knowledgeable colleague” who can respond faster, read more and crunch more data than humans.

As a result, Watson can play an important role in fighting cancer simply by doing all the heavy lifting of clinicians — crunching patient genomic data and sifting through hundreds of medical journals to see what treatment options might be available. As  Lukas Wartman of the Washington University’s School of Medicine points out, “I hope it will be possible for oncologists like me to quickly mine insights from the immense amount of genomic data that’s becoming available about individual patients by using Watson to identify potential drugs that target our patients’ specific genetic profiles.”

The hope is that this new Watson Genomics initiative will be about more than just enabling researchers at medical institutions to mine a much larger store of data. The real hope is that cognitive computing also implies creative computing – a computer that thinks out of the box based on industry best practices to arrive at solutions and recommendations that would impress even a human doctor.