Refinement of the monitoring of rainbow trout exposed to diseases

Kurt Buchmann

Aquaculture is at present providing more than 50 % of fish used for human consumption, and the expansion of the production is expected to continue during the next decades. However, diseases represent the most significant ongoing threat to aquaculture economy, sustainability and fish welfare regardless of geographical location, species, or production level. This is the background for conducting research elucidating health promoting and disease controlling methods in fish. With regard to rainbow trout research it is clear that studies on selective breeding and vaccinology are the most promising subjects at present.

When doing so it is necessary to expose rainbow trout to the pathogen in question and follow the development of disease in the different groups. We currently follow exposed fish around the clock and every second hour we evaluate any disease progression and remove fish with clinical signs from the tank (for euthanasia). This procedure reduces suffering of experimental fish. However, this is an extremely laborious process, which we seek to refine and to make it easier to detect sick fish at the earliest time after infection. This will secure that fish do not suffer unnessesarily during the studies. In this project a consortium of researchers from the University of Copenhagen and The Technological Institute (Teknologisk Institut) will therefore develop an automatic survey system based on video-surveillance and computer vision with artificial intelligence to send an alarm to the observer, whenever the clinical signs appear. This computer vision-based fish behavior analysis includes continuous, non-intrusive operations and will have a much higher resolution compared to methods used today.


Sign up for our newsletter - don't miss information about our annual symposium etc.