Some software errs on the side of caution too often, it is claimed
Katyanna Quach Tue 12 Jan 2021 // 08:33 UTC Share
Copy
A number of AI programs trained to detect diabetic eye damage struggle to perform consistently in the real world despite apparently excelling in clinical tests, say scientists in the US.
Academics led by the University of Washington School of Medicine tested seven algorithms from five companies: Eyenuk and Retina-AI Health in America, Airdoc in China, Retmaker of Portugal, and OphtAI in France. All of the models have gone through clinical studies, and are used – or can be used –to diagnose diabetic retinopathy, a complication of diabetes that damages blood vessels in the eye, leading to impaired vision or blindness.
Loading video.
VIDEO: JACR Deputy Editor-in-Chief Christoph I. Lee, MD, MS, and guest editor Reed A. Omary, MD, MS, discuss the JACR provocative special issue. view more
Credit: Journal of the American College of Radiology
Philadelphia, January 12, 2021 - A special issue of the
Journal of the American College of Radiology (JACR), published by Elsevier, challenges conventional wisdom across the imaging community. This collection of articles, the Provocative Issue, presents extreme opinions on pressing issues confronting radiologists with the deliberate aim of sparking positive dialog and debate that will lead to innovative solutions to improve patient care and imaging-related outcomes.
The issue is guest-edited by:
Should People Be Paid to Get a COVID-19 Vaccine? medscape.com - get the latest breaking news, showbiz & celebrity photos, sport news & rumours, viral videos and top stories from medscape.com Daily Mail and Mail on Sunday newspapers.
K-State medical director contributes to research behind updated CDC quarantine guidance
Monday, Jan. 11, 2021
MANHATTAN Kyle Goerl, the medical director of Kansas State University s Lafene Health Center, is part of a collaborative team that is providing research-based guidance during the COVID-19 pandemic. The team s latest research contributed to the updated quarantine guidance from the Centers for Disease Control and Prevention.
In the publication, Goerl and collaborators describe findings among a sample of COVID-19-exposed collegiate athletes in 17 states from June to October 2020. Twenty-five percent of the athletes tested positive during quarantine and the positive test occurred an average of 3.8 days after their quarantine started.
Sustained immune response memory detected in recovered COVID-19 patients
Since the coronavirus disease 2019 (COVID-19) pandemic began, scientists have been keen to know if primary infection in recovered individuals yields long-lasting immunity against the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), the virus that causes COVID-19.
A team of researchers at the University of Washington School of Medicine has found that people who had mild COVID-19 developed a multi-layered immune response against SARS-Cov-2. The researchers detected neutralizing antibodies against the virus for at least three months after recovering from the infection.
Specifically, the team found virus-specific memory B and T cells displaying anti-viral immunity. Memory B cells (MBCs) increase in number and express antibodies capable of neutralizing SARS-CoV-2.