How Microsoft is using AI to develop personalised healthcare

Elly Yates-Roberts
By Elly Yates-Roberts on 28 December 2018
How Microsoft is using AI to develop personalised healthcare

Microsoft is using artificial intelligence (AI) technology to improve healthcare and treatment processes. “Healthcare is possibly the biggest opportunity of all for machine learning. But it’s a very, very difficult field in which to work…,” said Christopher Bishop, director of the Microsoft Research Lab in Cambridge, in a recent episode of The Microsoft Research Podcast. “[It] is still in a relatively primitive state compared to some areas of manufacturing or other sectors.”

Bishop went on to discuss the ongoing AI revolution, explaining the benefit of machine learning; the revolutionary concept of enabling machines to learn from experiences in order to acquire their own intelligence.

Bishop suggests that using AI in healthcare could be “phenomenal” in terms of societal benefit. “By analysing data from millions of people, we have the potential to create personalised health solutions for each of those people as individuals,” Bishop said. When asked if machines could put medical staff out of work, he explains that they are excited to use AI. It can do the tedious and time-consuming parts of their job “much faster than the human with less variability [and] more accurately,” he said. It can also free up their time to do things that machines aren’t good at.

The fact that the healthcare industry is still in the early stages of digitalisation is of course a challenge facing the implementation of AI in this industry, but the innovation itself also poses a challenge – it requires that machines learn about the patterns that occur in large numbers of people by using their data. “In just the same way that a machine can learn your preferences for movies, it can learn what would be an appropriate course of treatment for you as an individual,” said Bishop.

To securely use people’s data without risk, the Microsoft Research Lab in Cambridge has been focused on confidential machine learning – a method of keeping data encrypted even whilst it is being processed in order to fuel machine learning. The technology, which has been deployed on the Microsoft Azure cloud, only allows the data to be decrypted once it is inside secure enclaves. “These are very tightly controlled software environments, protected with certain hardware technologies, making them very secure and meaning that only those with access to the keys to the data could ever access the data itself, even when it’s being processed,” said Bishop. “Even to the extent that Microsoft itself can’t access the data of its customers if it’s being decrypted inside these secure enclaves.” This is essential when working in an industry like healthcare as it can protect “the most valuable kinds of data.”

Bishop discusses the importance of finding a balance between how to use data and how to protect it. “At one extreme, we don’t want a free-for-all where data is readily available to everybody when it’s clearly private,” he said. “On the other hand, we don’t want to miss out on the enormous opportunity to improve lives and save lives that could come through applying machine learning in fields like healthcare.”

Listen to the Microsoft Research Podcast in full. 

 

Number of views (2694)/Comments (-)

Tags:

Comments are only visible to subscribers.