With 91% sensitivity but only 25% specificity, the tool is worthwhile for clinicians who remain wary of frequent false positives that would send patients with no fractures for unneeded imaging.

The ambitious project seeks to exponentially increase new data within the U.K. Biobank, a prominent large-scale biomedical database and research resource. 

A machine learning system has come along that needs no human labeling of data for training yet matches radiologists at classifying diseases on chest X-rays—including some that the model was not specifically taught to detect.

The acquisition represents a significant advance in interoperability and will combine networks that total over 80 billion images globally, including data from institutions such as Mayo Clinic, Cleveland Clinic and New York Presbyterian Hospital.

With the software’s help, the ratio of CTs requiring radiologist review to missed findings identified was 10:1, experts shared, adding that without the help of AI that ratio would be at least 66:1. 

Suggesting the move will significantly advance radiology’s specialtywide imperative to “ditch the disk,” Montreal-based Intelerad Medical Systems has announced it is acquiring a longtime competitor in the image-exchange space.

Denoising using deep learning techniques can boost the performance of 1.5T MR brain imaging, resulting in quality comparable or superior to 3T imaging. 

In this cohort, experts found that just 34% of surgeries were performed on symptomatic patients, which shifted their focus to other potential modes of detection.