A little after midnight of Dec 30, 2019: BlueDot, a Canadian AI start-up, raised an alarm about a disease outbreak in China. In the months that followed, authorities the world over paid heavily for neglecting this clarion call. When AI speaks, you must listen.
Wuhan - The onset of COVID-19 - Two healthcare workers under 30 years of age catch the coronavirus. One survives. Other succumbs.
In almost half-a-year of its existence, the coronavirus has equally baffled authorities and healthcare professionals. It is not just another flu. Even the last flu left us scrambling for a vaccine for almost 100 years.
Fortunately, this time we may not have to wait that long. Artificial intelligence , with its fast and strong strides, has offered tremendous hope.
Here’s a look at the remarkable progress it has made in a pithy span of just six months:
A fresh crop of applications and models appeared almost by midnight across the globe, most of them driven by start-ups and true to their salt. It’s difficult to list them all, but here are a few prominent ones:
Chan Zuckerberg Biohub built a model to know the number of undetected cases and their consequences on public health.
Closedloop, an AI start-up, created an “open-sourced vulnerability index”. The model identified patients who were at the most risk to develop severe complications from COVID-19.
Clevy.io, a French start-up, created a chatbot to channel COVID-19-related government communications. Today, it answers almost 3 million public queries every day.
While most researchers are using AI with imaging to identify patients with COVID-19 infection, a few are extending its use to identify the cases that will need extensive care in future. UC San Diego Health had engineered a machine learning algorithm to detect pneumonia, which is assisting in managing COVID-19 cases.
Two technologies are helping big time:
According to findings published in Nature Medicine, AI tools can quickly identify COVID-19 from the chest CT scans and the patients’ clinical data.
Researchers at Massachusetts General Hospital and Harvard Medical School are developing a deep learning algorithm to identify COVID-19-linked pulmonary disease likelihood and severity.
GE Healthcare recently announced a partnership with National Consortium of Intelligent Medical Imaging (NCIMI), UK to create an algorithm for predicting the complications, severity and long-term impact of COVID-19.
University of Copenhagen is creating models to estimate the need for intensive care units for patients.
Case Western Reserve University is using imaging to identify minute details in chest scans that cannot be detected by the human eye.
John T McDevitt, professor at NYU College of Dentistry and NYU Tandon School of Engineering, and his colleagues are working on an app that uses blood biomarkers and AI to determine the disease’s severity. With details like age, sex and blood biomarkers, the model can arrive at the case severity outputting numerals ranging from 0 to 100, zero being mild to hundred being critical.
Aside from directly preventing, detecting and managing the disease, AI is offering support on other fronts too. Here’s a look at the ones that have garnered most attention:
BenevolentAI, a UK based set-up, is using its platform to understand the impact of COVID-19 on the patient’s body. Its machine learning platform is helping to identify drugs that can inhibit the progression of the disease.
Baricitinib, a drug now in late-stage clinical trials as a potential cure for COVID-19, was discovered by BenevolentAI.
AWS-powered CORD-19 Search can help researchers quickly find COVID-19-related studies and documents. The website uses machine learning to identify relevant papers from its database of more than 128,000 research papers.
Both University of Copenhagen and Case Western Reserve University are using past data and patterns to suggest if a patient will need a ventilator and for how long.
Frontiers in Public Health journal published a paper titled COVID-19 Patient Health Prediction Using Boosted Random Forest Algorithm that will combine “healthcare data and demographic processing models with CXR scanning models.”
“AI has huge potential for analyzing large amounts of data quickly, an attribute that can have a big impact in a situation such as a pandemic.” Zahi Fayad, Director at the Icahn School of Medicine at Mount Sinai
Two challenges are paramount:
# The prognostics of COVID-19 are trickier than what was previously thought. While the current diagnostic methods can identify an infected person, they fail to disclose the extent of infection and its impact.
# Shortage of healthcare facilities has exacerbated the impact. In countries like Italy and Spain as well as cities like New York and Houston, it has led to devastating effects.
With the recent developments in AI, we are on the verge of addressing the above two limitations.
One thing has become clear as day – In the battle with COVID-19, AI is our only hope.
The future is promising with conversational Ai leading the way. This guide provides a roadmap to seamlessly integrate conversational Ai, enabling virtual assistants to enhance user engagement in augmented or virtual reality environments.
This website uses cookies to enhance website functionalities and improve your online experience. By browsing this website, you agree to the use of cookies as outlined in our Privacy Policy.
