Grad taps technology to speed review of COVID-19 literature
As COVID-19 emerged earlier this year, specialists around the globe scrambled to study and document the confounding new virus. That scientific knowledge is now accumulating rapidly. Recently, more than 4,000 new studies were published in a single week.
While this ever-growing mountain of data is promising, the sheer volume presents challenges for the physicians and scientists who tap into this treasure trove to help them develop treatment protocols and vaccines, says immunologist and EVMS graduate Tayab Waseem, PhD.
“With the rapid pace at which the literature is advancing and evolving, the medical and scientific community is struggling to keep pace,” says Dr. Waseem, who studied under the guidance of Elena Galkina, PhD, Professor of Microbiology and Molecular Cell Biology.
To address the challenge, the White House Office of Science and Technology Policy issued a challenge to the machine-learning community to develop an automated literature evaluation toolset for the medical research community.
In addition to his talents as an immunologist, Dr. Waseem also is an expert in artificial intelligence (AI). When he learned of the opportunity, he joined with Kaggle, a Google subsidiary and online community of nearly five million data scientists and machine-learning specialists.
Dr. Waseem is now the AI-literature reviews projects lead. In that capacity, he is leading a group of over 150 medical and scientific volunteers across 30 institutes and five countries. The group is using the CORD-19 data set, considered the largest single collection to date with 128,000 published articles and preprints, including at least 38,000 published just this year.
“We are utilizing CORD-19 and advanced natural language processing to expedite the analysis of disparate volumes of research in real-time to aid in mitigation of COVID-19,” Dr. Waseem says.
The project has drawn widespread interest. Many of the questions the group is addressing are coming from the leadership of major medical institutes and journals. Dr. Waseem has written journal articles about his work and he has been quoted in stories in Nature and Science Magazine. A recent story in the New England Journal of Medicine Resident 360 publication also mentions his work.
Dr. Waseem credits EVMS faculty mentors for encouraging his interest in the application of technology to medicine and science.
“I was fortunate enough to have a couple of mentors throughout my time at EVMS who aren’t afraid to push the technological envelope,” he says. "Clinically, Dr. Jose Morey and Dr. Alan Wagner were always on the cutting edge of technology and seeking to drive innovation in their respective fields. They always encouraged me to take the big leaps necessary to change the way medicine is practiced and to follow my passions which lie at the intersection of technology and medicine. On the basic science side, Dr. Patric Lundberg was always looking to bridge AI and basic sciences and his expertise has been invaluable.”
During the course of the project, Dr. Waseem has worked with EVMS faculty, medical and graduate students, and alumni from both the School of Medicine and School of Health Professions.
As he develops tools to mine COVID-19 research, he continues to build on his education. He is a Public Policy Fellow at the American Association of Immunologists and a future physician (he is a member of the EVMS MD Class of 2023).
He envisions one day leading a Medical Innovation Lab focusing on AI and other emerging technologies in the field of medicine.
“During this pandemic we have had an unprecedented breaking down of knowledge silos,” he says, “and I urge the medical and scientific community to not go back to the status quo. Keep innovating, improving, and advancing.”
The consortium assembled to address this issue is the first of its kind combining leaders in the fields of clinical medicine, research, and artificial intelligence. Dr. Waseem sees it as a template for the future. He believes true innovation will always come from a multidisciplinary approach.
“Man plus machine,” he says, “is the future of AI in medical augmentation.”