Crowdsourcing—pooling together small amounts of ideas, money and expertise—has become the most popular way to advance visions that otherwise would never leave the drawing pad due to lack of resources. Just about every professional who needs to raise money is turning to crowdsourcing to launch their project. Initiatives benefitting from crowdsourcing span the spectrum from new music releases to social causes to medical procedures. Crowdsourcing has become the way to help innovators change the world.
A recent example of the impressive power of crowdsourcing is the data science competition sponsored by the online platform Kaggle. 504 international teams competed in two challenges hosted by Kaggle: one on detecting epileptic seizures and the other on predicting epileptic seizures. The winner of the prediction challenge was a team of scientists from America and Australia—including mathematicians and software engineers. They were provided datasets derived from intracranial recordings from dogs and humans with epilepsy.From these data sets, they developed a computer algorithm based on electrical activity in the brain that can predict epileptic seizures within 82 percent accuracy. Now, in the true crowdsourcing tradition, their findings are already released so that researchers and health professionals can take advantage of the algorithms to bring about more effective relief to those suffering from epilepsy. Interestingly, no one on the team was a doctor. The Kaggle challenge was sponsored by the Epilepsy Foundation, American Epilepsy Society and the National Institute of Health’s Institute of Neurological Disorders.
Epilepsy is a brain disorder affecting approximately one percent of the population. Seizures occur in the brain, although they are visible in the body. It is the fourth most common neurological disorder and affects people regardless of age. The seizures are unpredictable and lead to other health problems over time. The most serious issues associated with epilepsy are the stigma attached to the disorder and the limited number of treatment options. Drugs are available, but they have harsh side effects. Surgery is the other option, where electrical devices are implanted in the brain to prevent seizures. Neither approach is producing truly satisfactory results, for the patient or in the eyes of healthcare providers. Hopefully this new prediction model will enable the development of more efficient treatment protocols.