Fifty million people have epilepsy, and many suffer brutally unpredictable seizures. If the seizures could be anticipated, electrical brain implants could stop a nascent seizure in its tracks, or at least give patients warning.
For 15 years the National Institutes of Health and other agencies spent $40 million trying — and never quite succeeding — to devise algorithms that could spot a growing neurologic storm. Then a frustrated group of epilepsy physicians invited computer nerds around the world to take a shot instead, providing data sets recorded from the brains of human epilepsy patients and epileptic dogs.
Three months — and $30,000 of prize money — later, the winners had produced algorithms predicting seizures with better than 84 percent accuracy.