Development of an interpretable artificial intelligence tool for colic detection in horses

For owners of high-value horses (>$100K), colic is one of the biggest challenges, as it can cause a horse to die within days. SolR Consulting has developed a system that can measure and archive a horse’s hydration history as well as reports of apparent state of health completed by the staff who care for it. This research project aims to use this data to automatically detect colic in a horse. To do this, a new artificial intelligence technology will be used that has the property of being interpretable, meaning that the learnings achieved are in the form of easy-to-understand rules. This will allow a horse owner to understand the reasoning behind the detection of potential colic to help them in their decision. In addition, being an artificial intelligence technology, this detection tool will continually refine itself over time and with the accumulating data. With this project, SolR Consulting will be able to complement their service offering and significantly increase their added value, allowing them to attract many more clients.

Faculty Supervisor:

Richard Labib

Student:

Partner:

SOLR Consulting Inc.

Discipline:

Mathematics

Sector:

Professional, scientific and technical services

University:

Polytechnique Montréal

Program: