Microsoft Research Teams Leverage AI And Machine Learning To Help Eradicate Cancer

Let’s face it, cancer sucks! We’re sure that at all of us knows someone who is battling cancer or has lost a friend of close family member due to the perils of cancer. Given that this disease touches just about all of us, it should come as no surprise that billions of dollars are poured into researching cancer and ways to eradicate it for good.

Luckily, the fine folks at Microsoft are well on their way to cracking the secret code that allows cancer to propagate in a host body. Microsoft’s researchers from around the world are working together using artificial intelligence to effectively “solve” the cancer problem.

Microsoft is attacking cancer from multiple fronts, with some of its research teams using machine learning and natural language processing to develop individualized cancer treatments for patients. Another team is using machine learning in conjunction with computer vision to give radiologist a faster way to model tumors inside and patient and determine a proper course of action for treatment.

microsoft 1

Take the latter team for example. Radiologists today must scan a patient’s tumor and 3D map it by hand before treatment can take place. This process can take up to four hours in some cases. However, this particular Microsoft research team has built an algorithm that can map a tumor in just minutes, which allows for even quicker treatment for patients.

Microsoft also has a team work on a “moonshot” project that “could one day allow scientists to program cells to fight diseases, including cancer.” Using DNA as a basis, Microsoft’s team is working to develop molecular computers that on their own can sniff out cancer cells and kill them.

“If we want to be able to program biology, then we actually need to be able to understand what it is biology computes in the first place,” said scientist Sara-Jane Dunn. “That is where I think we can have some major impacts.”


Andrew Phillips, the leader of Microsoft’s Bio Computation Group, adds, “It’s long term, but I think it will be technically possible in five to 10 years time to put in a smart molecular system that can detect disease.”

Helping along the way is Project Hanover, which uses machine learning to do the grunt work of “sifting through millions of pieces of fragmented information” to find the proper course of action for a patient’s cancer diagnosis.

“If we can use this knowledge base to present the research results most relevant for each specific patient, then a regular oncologist can take a look and make the best decision,” said Microsoft principal software architect Ravi Pandya.

But of course, Microsoft isn’t the only company that is looking to use artificial intelligence to help diagnose and fight cancer. Microsoft finds itself surrounded by Google’s DeepMind and IBM’s Watson which are also looking to better help doctors understand the inner working of what makes cancer “tick”.

“We are at this tantalizing moment where we’ve caught a glimpse of this really promising future, said Microsoft researcher Hoifung Poon. “But there is so much work to be done.”