AI algorithm IDs abnormal chest x-rays with 90% accuracy
Artificial intelligence (AI)-focused startup Qure.ai published a study Tuesday, Sept. 18, validating its chest x-ray algorithm trained on 1.2 million chest scans and radiology reports.
The deep-learning model—qXR—proved 90 percent accurate in detecting abnormal chest x-rays along with calcification, fibrosis, opacity and other abnormalities. The research was published in Cornell University’s online library for research, arXiv.org.
"This study is a series of exciting firsts," said Prashant Warier, CEO and co-founder of Qure.ai, in a statement. "This is the largest training dataset ever for a chest x-ray AI. It's also the largest validation study to date, measured against 2,000 x-rays—each read by three radiologists.”
According to the release, the platform also screens for tuberculosis and is used in public health screening programs across the globe.
"This is an exciting time for deep learning technologies in medicine," Warier added. "As these systems increase in accuracy, so will the viability of using deep learning to extend the reach of chest x-ray interpretation, improve reporting efficiency and save lives."
Matt joined Chicago’s TriMed team in 2018 covering all areas of health imaging after two years reporting on the hospital field. He holds a bachelor’s in English from UIC, and enjoys a good cup of coffee and an interesting documentary.