A clinical cloud-based deep learning in healthcare

The FDA’s approval of the first cloud-based machine learning application signals a major stride forward for artificial intelligence (AI) and machine learning in healthcare. The medical imaging platform, developed by Arterys, aims to be used in clinical settings to help physicians understand how a heart is functioning and to provide a precise assessment and diagnosis of the heart with accurate measurements of the volume of blood in each ventricle.

The approval was granted based on the platform’s ability to provide results as accurately as qualified professionals in the field. The key difference lies in the time required for analysis -- Arterys cuts examination time from up to an hour to just 15 seconds per case. “This is a huge deal – it’s the first time this new way of imaging has been cleared for clinical application. It’s about truly helping clinical workflow to move into the cloud and deep learning and do something pretty substantial. It opens the seals, and sets a precedent for what can be done,” said Fabien Beckers, one of the founders of Arterys to Forbes.

The platform, called Arterys Cardio DL, has already learned from 1000 cases and will continue to improve its knowledge as it examines more cases. Its goal is to become able to recognise and identify problems with minimal or no manual interventions.

This story was originally published in the Global Health and Travel issue of October 2017