An AI to alleviate the COVID-19 crisis? Thales & NEHS DIGITAL collaborate to form radiology solution targeted at hospitals
The COVID-19 pandemic has undoubtedly had an important impact on healthcare services worldwide, to the extent that hospital departments have become overcrowded. So, how can we help regarding this?
Thanks to the support from the Defense Innovation Agency, Thales & NEHS DIGITAL have joined forces to create an AI*-based solution that will allow healthcare professionals to have a first pre-diagnosis thanks to an analysis of lung images from the scanner. This project will benefit from Thales' long term expertise in radiology and digital technologies such as AI, and the solid establishment of NEHS DIGITAL** who is at the heart of French hospitals with dedicated solutions for telemedicine.
The solution, developed as part of a call for projects supported by the French Ministry of Armies, is set to deploy at the autumn 2020, and aims to improve patient care by enabling medical teams to perform better patient management and prioritise cases according to actual medical urgency. Using the FIDAC*** database from various hospitals, Thales will implement an infrastructure to train these algorithms through automatic learning, and will subsequently develop an AI that will initiate, in record time, automated diagnosis recommendations based on CT-scan images. Installed in hospitals thanks to NEHS’ solutions large coverage in France, this will help radiologists in hospitals, who are challenged with the responsibility of interpreting multiple images for each patient in a limited time to assess lung damage.
This large-scale project truly does set a precedent for managing future pathologies, as it is one of the first steps in supporting the practice of radiologists with AI in the context of COVID-19. It could be replicated for other pathologies in the near future!
* Artificial intelligence
** NEHS DIGITAL: subsidiary of the NEHS group
*** FIDAC project: database of chest CT-Scans from patients diagnosed with COVID-19, led by the SFR (French Society of Radiology) and supported by NEHS DIGITAL