DigiHeart: Machine learning applied to predicting and preventing production loss in aquaculture
The background for the Digiheart project is that up to 20% of farmed salmonids die every year prior to slaughter. Typically, episodes of mortality happen during stressful incidents and interventions such as parasite treatment, deteriorating water quality, and transport to slaughter. Currently, the causes of stress-related mortality are uncertain, but an increasing body of evidence suggests that the majority die from heart failure due to abnormal heart shapes or other types of heart disease.
Since cardiac morphology is a novel and promising indicator of mortality-risk, DigiHeart will work to understand the causes and consequences of deviating heart morphology during freshwater production and performace at sea. In praxis, the project will collect and measure salmon hearts prior to sea transfer, after sea transfer, and close to slaughter in order to develop a standardized sytem for on-farm evaluation of cardiac morphology and health. The morphological data will be combined with probabilistic causal models built using farm data to try and to predict if 1) various production parameters lead to deviated hearts and 2) if deviated hearts predict mortality related to stressors such as transprot to sea and delousing. The ultimate goal is that the results of the DigiHeart project will lead to a easy-to-use software tool that farmers can use to predict mortality-risk during routine operations.
Read more about the project here: https://www.nordforsk.org/projects/machine-learning-applied-predicting-and-preventing-production-loss-aquaculture-digiheart
The project kick-off was hosted by Fiskaaling and took place at the Nordic House, Tórshavn, Faroe Islands in late October 2021. https://www.fiskaaling.fo/t%C3%ADðindi/granskingarverkaetlanin-talgilt-hjarta-fer-av-bakkastokki/?lang=EN
Project duration: 2021-2024
Consortium: Norwegian Univeristy of Life Sciences (Ida Beitnes Johansen, Project Leader), DNV, Nofima, Aqua Kompetanse AS, University of Oslo, University of Göteborg, Swedish University of Agricultural Sciences, Fiskaaling
Faroe Islands- Bakkafrost, MOWI, Luna/HiddenFjord
Norway- AquaGen, Ellingsen Seafood and Midt-Norsk Havbruk AS
Nordforsk through the Nordic Research and Innovation Programme for Sustainable Aquaculture and internally by Fiskaaling.
Contact: Amanda Vang