Covid builds Facebook AI to predict possibility of worsening of symptoms –

Covid builds Facebook AI to predict possibility of worsening of symptoms

43 year old Dr. Dan Ponticello and 40-year-old Dr. Gabriel Gomez instigated a coronavirus disease (COVID-19) patient on January 8, 2021 at the COVID-19 ICU at Providence Mission Hospital in Mission Weijo, California.

Lucy Nicholson | Reuters

Facebook’s Artificial Intelligence researchers claim they have designed software that can predict the likelihood of a Kovid patient deteriorating or oxygenated based on their chest X-rays.

Facebook, which works with academics in NYU Langone Health’s Predictive Analytics Unit and Radiology Department on research, says the software can help doctors avoid sending patients at risk home early, while seeking oxygen in hospitals Is also helping.

The 10 researchers involved in the study – five from Facebook AI Research and five from NYU School of Medicine – have developed three machine-learning “models” in total, all slightly different.

One tries to predict the patient’s deterioration based on a single chest X-ray, another does the same with a sequence of X-rays, and the third uses a single X-ray to predict this. Know how much supplemental oxygen (if any) a patient may need.

The author said in a blog post published on Friday, “Our model using sequential chest X-rays can predict up to four days (96 hours) if a patient needs a more intensive care solution by human experts Might. “

William Moore, a professor of radiology at NYU Langone Health, said in a statement, “We are able to show that with the use of this AI algorithm, a serial chest radiograph, predicts increased need for care in patients with Cidid-19. can do “

He said: “As Kovid-19 is a major public health issue, the ability to predict the patient’s need for upgrading care – for example, ICU admission – will be essential for hospitals.”

To learn how to make forecasts, the AI ​​system was fed two datasets of non-Kovid patient chest X-rays and of 26,838 chest X-rays out of 4,914 Kovid patients.

The researchers said they used an AI technique called “motion contrast” to train a neural network to extract information from X-ray images of the chest. A neural network is a computing system vaguely inspired by the human brain that can spot patterns and recognize relationships between vast amounts of data.

The research was published this week by Facebook, but experts have already questioned how effective AI software can be in practice.

“From a machine learning perspective, one would need to study how well this translates to new, unseen data from various hospitals and patient populations,” said Ben Gawker, who via email at Imperial College London Imaging Researches machine learning for. “From my skim readings, it appears that all the data (training and testing) is coming from the same hospital.”

Researchers at Facebook and NYU said: “These are not model products, but research solutions, intended to help hospitals in the coming days and months with resource planning. While hospitals have their own specific data sets, often they have Are not. The computational power required to train deep learning models from scratch. “

“We are open-sourcing our professed model (and publishing our results) so that hospitals with limited computational resources can fix the model using their own data,” he said.


Leave a Reply

Your email address will not be published.