A 17-year-old girl, Brittany Wegner, has devised the least invasive way to assess the presence of breast cancer from tissue samples. The technique uses a cloud-based neural network (or an artificial ‘brain’) to analyze tissue samples for breast cancer. The project brought Wegner the first prize in this year’s Google Science Fair.
This cloud based neural network developed by Brittany Wegner from Florida can accurately examine tissue samples for signs of breast cancer. Wegner is trying to detect breast cancer well ahead it reaches a critical stage. “That is what I’m trying to do with my neural network,” she said.
Basically, this artificial neural network is a computer program similar to our brain structure. However, neural networks can be built to handle more complex patterns, compared to human brain.
Wegner built her neural network with Java programming language, and then deployed it in a cloud based platform. She ran 7.6 million trials on it, and found it is 99.1 percent sensitive to malignancy. Wegner compared her custom-built network to three off-the-shelf products that she tweaked to work with her dataset. She found that her sensitivity to malignancy was 4.97 percent better than the commercial networks.
For this breakthrough invention in the cloud-based neural network, Brittany Wegner won this year’s Google Science Fair grand prize – $50,000 in scholarship money, an internship with a fair sponsor and a 10 day trip to the Galápagos Islands. Wegner plans to deploy her design in hospitals. She also wants to extend it to other types of cancer detection.