Our research focuses on developing algorithms to automatically extract pertinent information from such notes on a large-scale and in real-time to support secondary use of narrative data for canine health surveillance. We will develop bespoke methods to automatically ‘read’ narratives and extract mentions and values of key observable clinical signs and variables (e.g. temperature). That data will be normalised and then used for disease surveillance that is being built by our colleagues in Bristol, Liverpool and Lancaster.
We are both excited and determined to make veterinary narrative accessible for large-scale analyses, integrate it other veterinary data and thus unlock the vital information that is stored in free text so that we can improve our understanding of canine diseases and their management.Professor Goran Nenadic, Professor of Computer Science at the University of Manchester