• Reduce text

    Reduce text
  • Restore text size

    Restore text size
  • Increase the text

    Increase the text
  • Print


Fish carcass quality: imaging at the service of more efficient breeding schemes

From the economic and environmental points of view, improving processing yields is an important challenge in fish farming.  Coupling ultrasonography and digital photography enables the measurement of internal and external morphological parameters and the prediction of processing yields on live fish.  Selection achieved directly using these fish should be optimised.

Représentation de la conformation type des lots extrêmes en rendement carcasse étêtée (en rouge : fort rendement, en bleu : faible rendement). © INRA, Jérôme Bugeon
Updated on 01/15/2015
Published on 12/09/2014

The genetic improvement of fish based on traits such as processing yields (gutting, filleting) is currently based on measurements on slaughtered relatives, which limits the efficiency of selection. The development of measurement methods that can be applied on live animals is an important challenge to improve selection. Because processing yields are linked to the relative distribution and development of muscle, bone and adipose tissue masses, morphological analysis both internally (2D ultrasound) and externally (digital imaging) constitutes a promising path to predict yields.

 Two coupled imaging techniques

 Two imaging techniques were coupled to obtain measurements on 2000 rainbow trout (600 families).  Muscle wall thickness and the depth of the abdominal cavity were measured using 2D ultrasonography.  The collection of 19 morphological points on digital photos was used to measure body lengths and surface areas. All these data made it possible to establish equations for yield prediction (carcass, head, headless carcass).

 Possible improvements to selection for processing yields

 An estimation of genetic parameters for these predictors, and a comparison of predicted and actual processing yields, were carried out. Although the heritabilities (h²) of the predicted yields were generally half those of the yields  measured (h² 0.25-0.28 versus 0.47-0.55), some very good genetic correlations (rG) were observed between the two measurements (rG 0.87-0.90), which rendered them good candidates as selection criteria.

 Simulations of potential gains by generation showed that in the case of an improvement scheme where candidates were selected based on the predictive model (10% pressure), the gain achieved was 0.85 improvement in headless carcass yield versus 0.7 with the current scheme (40%  pressure on relatives).  This difference could be explained by the greater selection pressure permitted using predictors measurable in vivo.  Use of these imaging methods could thus permit the optimisation of selection by indexing candidates on the predicted value of their yields.

 Some perspectives

 The predictive equations still need to be improved to enhance their accuracy.  The collection of external morphological points in 3D rather than 2D would enable measurement of the share of variability in muscle growth in terms of thickness and thus improve the prediction of processing yields.

 These simulations need to be validated by actual selection programmes based on these indirect criteria (European FISHBOOST project); the transfer of these results to other farmed fish species will also be studied.

Scientific contact(s):


Grant report FranceAgrimer n°083/08/C et FEP n°30900/2009.
Pierrick Haffray, Jérôme Bugeon, Quentin Rivard, Benjamin Quittet, Sophie Puyo, Jean Michel Allamelou, Marc Vandeputte, et Mathilde Dupont-Nivet. Genetic parameters of in-vivo prediction of carcass, head and fillet yields by internal ultrasound and 2D external imagery in large rainbow trout (Oncorhynchus mykiss). Aquaculture 410-411:236-244, 2013.

Pierrick Haffray, Jérôme Bugeon, Quentin Rivard, Benjamin Quittet, Marc Vandeputte,  Mathilde Dupont-Nivet. Selecting for processing yields: genetic parameters of non lethal indirect criteria in rainbow trout Oncorhynchus mykiss. 11. International Symposium on Genetics in Aquaculture, Auburn (US) 2012/06/24-30, 37 (Oral communication – abstract)