This teenage girl teaches computers to spot cancer

This week, Florida teenager Brittany Wenger is in London to speak to the Royal Society of Medicine about her artificial intelligence software, which could help doctors to diagnose breast cancer more quickly and easily, the BBC reports.

The 19-year old built the prototype software after her cousin was diagnosed with breast cancer. It’s a type of neural network, software that’s programmed to mimic the human brain which can become more accurate over time. The aim is to “teach” computers how to detect patterns in breast cancer biopsies.

The programme is currently being tested in two hospitals, in Italy and the US, and Brittany has been invited to demonstrate her research at the White House as well as winning a $50,000 university scholarship from Google.

She’s now studying biology at Duke University in North Carolina and wants to become a paediatric oncologist and research scientist. She’s also working on her next project: a cloud-based computer program for leukaemia, which will look for genetic patterns that could predict a relapse.

And she’s not the only teen genius trying to improve the speed of cancer diagnoses. In 2012, 15-year old Jack Andraka won Intel’s International Science and Engineering Fair with a paper sensor to detect pancreatic cancer that’s 168 times faster and 26,000 times less expensive than standard tests. And this year, 15-year old Nathan Han from Boston won the same prize, worth a hefty $75,000, for software that evaluates mutations in the BRCA1 gene (which increases the risk of breast and ovarian cancer) to see how harmful they are. His program has an 81% accuracy rate, compared to 40% for current tests.

Image via Ed Uthman’s Flickr.