The Associate Dean for the Master of Business Analytics at the Massachusetts Institute of Technology (MIT) Dimitris Bertsimas and his team of students have created a website to rapidly develop and deliver tools for use by hospitals, government officials, and healthcare institutions in the United States to combat the spread of COVID-19.
The website offers predictions of COVID-19 infections, hospitalizations, and deaths in all the states of the United States. It is based on a new epidemiological model that was specifically built for COVID-19.
Speaking to Greek Reporter, Bertsismas says that the website data can help health authorities make decisions regarding vital questions of life and death.
“We are at the point in many parts of the world the resources available are below what is needed. So as a result we have make very tough decisions — and instead of making them arbitrarily we need to think about them rationally,” he states.
Bertsismas explains that in the US, and perhaps in other countries too, there maybe enough capacity in terms of ventilators as a whole, but they are not in the right place at the right time.
“Our model will optimize where to send (ventilators) either from the federal government or from other US states to be able to cover the demand and move ventilators where they are needed most.”
Bertsimas says that this has already happened. “Last Saturday, for example, Oregon sent 140 ventilators to New York. On the website we have a model of predicting the evolution of the disease in all states. The model can help in making allocation decisions.”
The Greek scientist and his group of approximately 20 graduate students in the Operations Research Center at MIT are now working with a number of hospitals to make rational allocation decisions.
“Hospitals may have enough resources as a whole, but specific hospitals in the system may be under. The question is how to optimize the totality of the resources by moving people, nurses and doctors, ventilators and potentially patients,” Bertsimas says.
There is also an interactive part of the website which helps people understand the correlations between disease and outcome.
“We try to help at the global level, at the state level or at the country level prioritizing resources and we also trying to help individual patients,” the researcher explains.
Bertsismas notes that currently a if a patient calls in with some symptoms, he or she is most likely to be asked to stay at home, unless he or she is very ill.
“This is not necessarily correct. Certain patients need to go to hospital. The advice should be more personal and based on data, not just an opinion. Somebody should be utilizing the data we have,” he urged.
The group is in the process of developing a personalized prediction calculator that assesses the probability of length of stay at the ICU and mortality risk using data it has received from major hospitals in Boston, Massachusetts, Wuhan, China and Cremona, Italy.
“We need personalized advice because this disease is going to be with us for some time,” says Bertsimas.
See all the latest news from Greece and the world at Greekreporter.com. Contact our newsroom to report an update or send your story, photos and videos. Follow GR on Google News and subscribe here to our daily email!
