Ann: MX1 - Appendix 4C and Quarterly Activities Statement, page-30

I'm interested in the article you posted as it doesn't relate to Micro-X's CT scanner - the scanner doesn't use AI to reconstruct a 3D image of the brain nor perform an initial diagnosis. they just use an image reconstruction algorithm similar in general terms to the ones used in conventional CT machines today, but different in that it takes into account the unique configuration of the emission and detector. The image is then sent to a radiologist, who can look at the image and know whether it is a stroke or a bleed - and give advice so that clinical care can be administered straight away if appropriate. The strength of this is that they've miniaturised the CT which is the clinical standard, if it meets this standard there's no desperate need for an overlaid AI classification model to decipher where the clot or the bleed is, it'll be to the standard that radiologists use already to tell the difference with the human eye. You can add a model over the top to augment our ability and make it better, but we aren't relying on any model to perform a diagnosis.

EMvision on the other hand do use an AI classification model to determine whether there is a thrombus or a haemorrhage from my knowledge.