Facebook is publishing research conducted by its artificial intelligence (AI) unit in an effort to help healthcare providers determine in advance if a coronavirus patient may need more intensive care solutions and adjust resources accordingly.
The research done in collaboration with NYU Langone Health’s Predictive Analytics Unit and Department of Radiology can help produce predictions based on chest X-rays.
Facebook in a detailed blog post recently noted that it developed two AI models, one based on a single chest X-ray, and another from a series of X-rays, that could help forecast if a patient infected by the coronavirus is likely to get worse. A third model predicts the amount of extra oxygen a COVID-19 patient might need.
Facebook’s AI models generally did a better job than a human when it came to forecasting up to four days in advance if a patient will need more intensive care resources.
“These predictions could help doctors avoid sending at-risk patients home too soon, and help hospitals better predict demand for supplemental oxygen and other limited resources,” read a statement by Facebook employees in the blog post.
The models, based on research carried out by Facebook AI researchers Anuroop Sriram, Matthew Muckley, Koustuv Sinha, and Nafissa Yakubova, have been published as open-source materials to assist the global healthcare community amid the ongoing outbreak of COVID-19.
In order to find out how to make predictions, the AI system was fed two datasets of non-Covid affected person chest X-rays and a dataset of 26,838 chest X-rays from 4,914 COVID 19 sufferers.
The researchers mentioned they used an AI method referred to as “momentum contrast” to prepare a neural community to extract data from chest X-ray pictures. A neural community is a computing system vaguely impressed by the human mind that may spot patterns and acknowledge relationships between huge quantities of knowledge.
This article was originally published by Mashable. Algoworks does not take any credit and is not responsible for the information shared in the article.