Our red dot® algorithm is based on state-of-the-art deep learning models. Thanks to their specific architecture and ability to model complex functions, deep learning models have proved to be extremely efficient at processing images. Our red dot® algorithm was trained on over 30,000 CXRs with detailed annotations from certified radiologists. The training process teaches the algorithm to both classify a CXR and localize its findings. This localization is rendered through heatmaps and allows to understand and interpret red dot® output.
How it works
Pioneering artificial intelligence in healthcare
Watch the video to see it in action
Using standard HL7 and DICOM messaging, the red dot® platform will retrieve, receive and process each CXR examination at the point of acquisition and send an electronic notification back to the Trust RIS (or PACS in a PACS driven reporting scenario) to indicate whether the examination is normal, or whether an abnormality has been indicated. This notification will prioritise the examination within the existing reporting worklists for urgent reporting. Messaging and CXR images are received via secure encrypted VPN to the red dot® platform, via the AI Gateway from the Trust RIS/PACS. All patient examination data resides in fully NHS accredited data centres