Is artificial intelligence about to revolutionise cardiovascular ultrasound diagnosis?

Artificial intelligence (AI) has transformed most areas of our daily life. Our searches online, our shopping choices, our transport methods, even the organisation of our factories are guided by machines that learn optimal patterns of behaviour to maximise sales or productivity. Can a similar revolution be unleashed within healthcare? Methods to optimise the organisation and […]Artificial intelligence (AI) has transformed most areas of our daily life. Our searches online, our shopping choices, our transport methods, even the organisation of our factories are guided by machines that learn optimal patterns of behaviour to maximise sales or productivity. Can a similar revolution be unleashed within healthcare? Methods to optimise the organisation and patient flows through hospitals, early warning monitoring signs for acutely unwell patients and advice on likely diagnoses are in development or being trialled. In addition, at the forefront of medical applications are techniques to handle medical images. This is largely because image recognition and handling has been fundamental to many of the recent applications of AI. Autonomous cars rely heavily on image recognition to navigate along roads, security checks at airports and on the street rely on facial recognition, and internet searches are driven by image similarity. The powerful computational approaches used to handle images can, therefore, be directed towards large medical imaging datasets. With accurate ground truth information about what is contained within the image and how this relates to different medical conditions, machines can be taught to recognise images, identify areas within images and report on what the image shows.

The content on this page is provided by the individuals concerned and does not represent the views or opinions of RAD Magazine.