Sep 19, 2016
Google isn’t the only tech giant hoping that artificial intelligence can aid the fight against cancer. Enter a Microsoft Research machine-learning project, dubbed Hanover, that aims to ingest all the papers and automatically process legions of biomedical papers, creating “genome-scale” databases that could predict which drug cocktails would be the most effective against a given cancer type.
Researchers at Oregon Health & Science University’s Knight Cancer Institute are working with Hanover’s architect, Hoifung Poon, to use the system to find drug combinations effective in fighting acute myeloid leukemia, an often-fatal cancer where treatment hasn’t improved much in decades. They include Jeff Tyner, and the institute’s director, Brian Druker, best known for pioneering Gleevec, a blockbuster drug for a different type of leukemia now owned by Novartis, that’s helped double those patients’ five-year survival rate since the 1990s.
Cancer is caused by genetic mutations that make cells grow and multiply out of control. Better ability to find those specific mutations has enabled new types of drugs that target the disease more precisely, raising the odds of survival. There are more than 800 medicines and vaccines in clinical trials to treat cancer, according to a 2015 report by Pharmaceutical Research and Manufacturers of America. At the same time, the rising speed and falling cost for sequencing genes has boosted research, the development of therapies and means more cancer patients can gain exact data on their case.
“In oncology, it’s almost impossible for us to think this is doable without machine learning”
Microsoft is also teaming with the Knight Cancer Institute on AI that would personalize those drug mixes on a patient-by-patient basis. They’re primarily focused on acute myeloid leukemia, where you might end up battling multiple leukemias at once — machine learning could identify just what you’re dealing with and treat it accordingly. Another effort would lean heavily on computer vision to understand how a tumor is reacting to treatments. Human doctors can easily identify tumors, Microsoft notes, but they can’t always tell how tumors are changing or how they’re affecting the health of nearby cells.
The Redmond crew is even more ambitious than that. It’s embarking on a “moonshot” that would rely on AI to program cells for fighting cancer and other diseases. You could create a “molecular computer” that could watch out for diseases and trigger a response, tackling the illness only when it reaches a given part of the body. Many current cancer treatments are indiscriminate, killing healthy cells alongside cancerous ones.
It’s a far-reaching initiative, and it may well help by making both sense of once-overwhelming amounts of data as well as developing more precise responses to cancer. However, it’s easy to understand if you’re skeptical. As cancer researcher Dr. Brooke Magnanti observes on Twitter, this comes across as the classic example of a tech company thinking that it can solve major societal issues by throwing enough data at the problem. Cancer is a “group of diseases” that you can’t easily fix, she says. She adds that modern computers are only faster at accomplishing tasks, not smarter — they aren’t likely to discover things that humans don’t already know. This isn’t to say that Microsoft’s efforts are a waste, just that it may be overly optimistic about its ability to thwart diseases that have stymied the medical field for decades.
Smart Screening, Intelligent Hiring, Proactive Alerts, Effective Compliance and Peace of Mind.