Newswise — Rockville, Md. -- Scientists have found that with minimal training, members of the general public may be able to match the accuracy of experts in interpreting medical images of the eye. The work, published in Translational Vision Science & Technology (TVST), introduces a world where individuals without medical training could contribute to decisions made in the clinic.
The paper, “The Accuracy and Reliability of Crowdsource Annotations of Digital Retinal Images,” describes results from over 5,000 internet users who volunteered to interpret 100 medical images. Optional training was offered, and those who pursued the training performed as well as medical experts in marking areas of the images that suggested a clinical issue.
“We were uncertain how accurate or effective [crowdsourcing] would be given the variation, subtlety and complexity of medical images,” said coauthor Danny Mitry, MD, PhD, FRCOphth, of Moorfields Eye Hospital, London.
The research team suggests the findings could help solve one of the biggest challenges in health care — a lack of medical professionals. Accurate crowdsourced medical image reading could cut costs and reduce the time between having an image taken and interpreted. Crowdsourcing can also facilitate telemedicine, where an image taken in a clinic can be interpreted by someone anywhere in the world.
“We plan to develop a more sophisticated and structured training module as well as a scoring system,” said Mitry. “We hope to test different and multiple imaging techniques in future studies.”
The ARVO journal Translational Vision Science & Technology (TVST), is an open access, online only, peer-reviewed journal emphasizing multidisciplinary research that bridges the gap between basic research and clinical care, available at tvst.arvojournals.org.
ARVO, an organization of nearly 12,000 researchers from over 75 countries, advances research worldwide into understanding the visual system and preventing, treating and curing its disorders. In addition to TVST, ARVO publishes the Investigative Ophthalmology & Visual Science and the Journal of Vision.
MEDIA CONTACTRegister for reporter access to contact details
Journal of Vision; Investigative Ophthalmology & Visual Science; Translational Vision Science & Technology (TVST)