How Data Science Can Help Cure Cancer

How Data Science Can Help Cure Cancer

By November 22, 2016 Booz Allen, Data Science No Comments

How Data Science Can Help Cure Cancer

I will never forget that call.image001

“Kelly has cancer,” my dad said softly.

Knees weak, I sat down on the bed. I didn’t know if my sister was going to live. And, despite us having spent decades doing everything together, she’d have to fight this battle on her own. I’m not the only one who’s heard that kind of call. The moment I experienced was not singular to me, it is one that is repeated over 12.7 million times each year – with over half of those ultimately not surviving.

People have been talking about curing cancer for decades, but today there’s reason for greater hope than ever before. For the first time, the technology for storing and analyzing data – and scientists who can find insights within that data – is so good that doctors and medical researchers aren’t the only ones who can help find a cure. Data scientists of all backgrounds are developing groundbreaking algorithms to detect early signs of disease.

Lack of available data is the biggest challenge in leveraging talent in the data science community to help address medicine’s biggest problems. HIPAA laws greatly restrict the storage, dissemination, and use of patient data. Data is the fuel of any analysis and, without it, inertia will reign supreme.

Despite generating more medical data than ever before, much of it digitized, most people can’t access their own health data much less data from a population large enough to reach any viable conclusions. This is a good thing, in some ways.  No one should have to question the privacy of this very personal data. That said, the rise of bioinformatics and the digitization of patient data offers an opportunity that can’t be ignored.

To fully arm data scientists to help fight some of these public health battles we should:

  • Establish a national database for the storage and dissemination of medical data including charts, results from laboratory tests, and medical images.
  • Include an opt-in clause on HIPAA forms allowing patients to volunteer to have their data anonymized, tagged with a UID to allow for longitudinal studies, and shared in the aforementioned database.
  • Host challenges that allow medical researchers to present problems that data scientists can compete to solve using patient-provided data in a manner similar to the Data Science Bowl.

In his last state of the union address, President Barack Obama put Vice President Joe Biden in charge of a “moonshot” to accelerate the fight against cancer. The initiative aims to make more therapies available to more patients and improve our ability to prevent cancer and detect it early. My sister was lucky enough to survive cancer; Biden’s son was not. Hopefully, by fighting these diseases as a community of data scientists, doctors, and medical researchers, we will help end the heartbreaking phone calls that change so many families’ lives.

—Written by Eric Druker