Using clinical notes to support canine health surveillance

Developing algorithms to enable the text mining of clinical notes from veterinary consultations, Dr Mercedes Arguello Casteleiro and colleagues at the University of Manchester are seeking to use free text narrative data to support canine disease surveillance.

Each veterinary consultation is documented with a clinical note typed by a vet, recording key reported clinical signs and discusses treatments. These narratives are often insightful but frequently aren't processed, since their free-text format makes it difficult for direct use in traditional disease surveillance methods. 

We're developing algorithms to automatically read and extract pertinent information from such notes on both a large-scale and in real-time. This includes extracting mentions and values of key observable clinical signs and variables, such as temperature.

The extracted narrative data will  be normalised and put to secondary use in supporting canine disease surveillance.