Artificial intelligence system for radiologists

December 7, 2017

DeepRadiology, a deep learning artificial intelligence medical company, has announces a novel CT head system with a performance that exceeds that of radiologists.

Artificial intelligence and machine learning are becoming commonplace in healthcare. Primarily these developments are marketed to help support the work of healthcare professionals in making decisions or acting as a means to verify something observed by the human medic. See, for example, the Digital Journal article “Digital First: Clinical transformation of pathology“; here we outline a number of digital and technological innovations shaping pathology services.

Radiology has so far been exempt from the smart technology process. This is set to change with the new technology from medical technology start-up DeepRadiology. The new system is designed to detect clinically significant pathologies in computerized tomography scans of the head. In trials the error rates are described as considerably superior compared with published error rates for professional radiologists. At the heart is a software system developed using the artificial intelligence techniques. The software was taught the knowledge contained in most major radiology textbooks plus the cumulative experience of some 9 million computerized tomography scan images, where the software was taught which answers were correct. The functionality of the system is described in a white paper titled “DeepRadiologyNet: Radiologist Level Pathology Detection in CT Head Images.” In terms of further developments, DeepRadiology is working on smart learning for images produced using X-rays, magnetic resonance imaging and ultrasound. These, together with the new artificial intelligence omputerized tomography scan, was presented at the annual meeting of the Radiological Society of North America, which took place in November 2017.

The views expressed in content distributed by Newstex and its re-distributors (collectively, “Newstex Authoritative Content”) are solely those of the respective author(s) and not necessarily the views of Newstex et al. It is provided as general information only on an “AS IS” basis, without warranties and conferring no rights, which should not be relied upon as professional advice. Newstex et al. make no claims, promises or guarantees regarding its accuracy or completeness, nor as to the quality of the opinions and commentary contained therein.

This article was from Digital Journal and was legally licensed through the NewsCred publisher network. Please direct all licensing questions to

Great ! Thanks for your subscription !

You will soon receive the first Content Loop Newsletter