A new algorithm has been developed that slashes the time it takes to assess medical scans.
In a paper by MIT researchers, an algorithm is presented that is 1,000 times faster at deciphering magnetic resonance imaging (MRI) scans, to identify brain tumours – that on average currently take about two hours.
“You have two different images of two different brains, put them on top of each other, and you start wiggling one until one fits the other. Mathematically, this optimisation procedure takes a long time,” said Adrian Dalca, senior author on the paper and postdoc at Massachusetts General Hospital and CSAI.
In tests, the VoxelMorph algorithm performed as well as traditional methods but much faster.
Our article, algorithms transforming our world, highlight a few of the algorithms being used to increase efficiency and accuracy in areas including energy, employment and natural disasters.