AUTHOR=Lindberg S-K. , Durland E. , Heia K. , Noble C. , Alvestad R. , Difford G.F. 

TITLE=Digital scoring of welfare traits in Atlantic salmon (Salmo salar L.) - a proof of concept study quantifying dorsal fin haemorrhaging via hyperspectral imaging

JOURNAL=Frontiers in Animal Science

VOLUME=Volume 4 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/animal-science/articles/10.3389/fanim.2023.1162384

DOI=10.3389/fanim.2023.1162384

ISSN=2673-6225

ABSTRACT=Morphological injuries are well-established Operational Welfare Indicators (OWIs) for farmed animals including fish. They are used in animal welfare audits and are often scored manually by human observers. The scoring of injury traits by human observers can be laborious, as well as prone to subjectivity and error.  In addition, operational injury scoring schemes can group mixed traits within each injury classification for both speed and ease of use in farming situations. This is challenging if an observer wishes to elucidate specific traits within the injury classification. The use of automated tools based on cameras and digital image analysis shows promise for non-invasive, reliable, and continuous welfare monitoring. In this study we evaluated the use of hyperspectral imaging systems to quantify the presence and severity of external haemorrhaging in Atlantic salmon focusing on dorsal fins as a proof of concept OWI. We compared injury estimates from these systems to manually scored visual assessments from two inexperienced human observers for a total of 234 post-smolt Atlantic salmon. The hyperspectral imaging platform was robust at detecting blood in fins and could help classify active injuries more accurately than human observers. The overall accuracy between the two methods and scoring systems was 84-94%. Classifying fin size/erosion severity, as assessed by observers (scale 0-3), was positively correlated to the amount of detected blood per unit surface area in the fin (fin haemorrhaging index) in general but with substantial variation in the amount of detected blood per fin size score class. Accuracy between the fin haemorrhaging index and the human observers was moderate (0.61 and 0.57) and on par with the agreement between the two human observers (0.68), demonstrating the difficulty in classifying injuries that result in a reduction in fin size but may or may not result in fin haemorrhaging. These results demonstrate the potential power of hyperspectral imaging to improve welfare audits in aquaculture, especially where manual injury classification schemes have potentially mixed traits, while simultaneously highlighting the need for further testing and validation to integrate these tools into existing welfare management programs.