A World First: Supercomputer Predicts Disease Before It Happens
The development of the supercomputer and trained algorithms continues. Finally, researchers have developed a system that predicts serious brain hemorrhages before they happen.
The Pawsey Supercomputing Centre, a supercomputing company located in Perth, Australia, provides services ranging from star mapping to weather forecasting and solving complex problems. Researchers decided to use the processing power here for the health sector this time and implemented the world’s first algorithm that predicts disease before it happens.
Disease Can Be Predicted 20 Minutes Before it Happens
Shiv Meka, head of data science at the Royal Perth Hospital, said more than 40,000 hours of patient data were collected from 200 patients to develop the algorithm. After working on 20 different data models, Meka and her team created a model that predicts whether brain haemorrhage disease will happen in the near future. The supercomputer-based model, whose results were extremely successful, can give consistent predictions up to 20 minutes before bleeding. Meka stated that this success would not have been possible without Pawsey, and said, “So you make sense of this data and then you train the model to predict an event.”
Pawsey Center’s Setonix supercomputer was recently named the 15th most powerful research supercomputer in the world. Setonix, Australia’s most powerful system for research, has been used to predict very serious brain haemorrhages.
Royal Perth Hospital intensive care specialist Robert McNamara said the project marked a turning point in the treatment of traumatic brain injuries. He added that it is common for medical personnel to be caught off guard in cases of traumatic brain injury, according to McNamara. Because it is known that while the brain activities and pressure of a patient in intensive care seem normal, the situation can change in just a few minutes.
A First in the World
The complex algorithm created allows researchers to monitor every “heartbeat, vital sign, and pressure wave” from the patient. “For the first time, we’ll be able to intervene before it happens,” said Meka and McNamara, who have created the world’s first operational traumatic intracranial hypertension prediction algorithm.
Mark Stickells, executive director of the Pawsey Supercomputing Research Center, said the algorithm is an example of how advances in data science and computing capabilities, developed by analyzing astronomical data, can help people and save lives. “Medical science is increasingly aware of developments in computing, machine learning and artificial intelligence.”