Communication Dans Un Congrès Année : 2025

Computed tomography for selection of anatomical sites to predict in vivo body composition in pigs and dairy goats

Résumé

Precise phenotyping of body composition and its temporal changes are essential in animal research and breeding, both for genetic selection through better characterisation of complex traits and for adapting breeding strategies to production goals and societal expectations. However, phenotyping in vivo body composition remains a challenge. Our objective was to determine anatomical sites from which in vivo whole-body composition of pigs and dairy goats can be predicted, based on computed tomography (CT) images analysis. Twenty-two crossbred Pietrain x (Large White x Landrace) gilts (152.5 ± 0.9 days old, 99.2 ± 7.5 kg body weight (BW)) and 20 dairy Alpine goats (3.0 ± 0.6 years old, 54.7 ± 6.6 kg BW) were scanned after anaesthesia using CT. Fat (FT) and lean thicknesses (LT) were calculated from 15 cm × 15 cm areas on the back of the animals, centred on the spine using vertebrae as landmarks. Linear regressions were calculated to predict CT fat and lean volumes of the whole animal, excluding offal, from either 2 predictors (BW and FT) or 3 predictors (BW, FT and LT). The best anatomical site to predict fat volume with 2 or 3 predictors is the 13th thoracic vertebra (adjusted R²: 0.70, coefficient of variation of the root mean square error (CVRMSE): 6.6%) in pigs and the 1st thoracic vertebra (adjusted R²: 0.57, CVRMSE: 22.7%) in goats. For the prediction of lean volume, the 1st lumbar vertebra with 3 predictors (adjusted R²: 0.59, CVRMSE: 3.7%) or the 5th thoracic vertebra with 2 predictors (adjusted R²: 0.45, CVRMSE: 4.3%) are the best anatomical sites in pigs. In goats, the best anatomical sites were either the 5th lumbar vertebra with 3 predictors (adjusted R²: 0.30, CVRMSE: 9.1%) or the 8th thoracic vertebra with 2 predictors (adjusted R²: 0.22, CVRMSE: 10.0%). The identification of anatomical sites that are representative of the whole body of the animal provides for facilitating in vivo phenotyping of body composition with less constraining methods than CT. This work is part of Cocoone project, funded by Carnot France Futur Elevage.

Fichier non déposé

Dates et versions

hal-05281501 , version 1 (28-10-2025)

Identifiants

  • HAL Id : hal-05281501 , version 1

Citer

Anne de La Torre, Couvert Chloé, Nathalie Quiniou, Mathieu Monziols, Ludovic Brossard. Computed tomography for selection of anatomical sites to predict in vivo body composition in pigs and dairy goats. 76. Annual Meeting of the European Federation of Animal Science (EAAP), Aug 2025, Innsbruck, Austria. pp.248. ⟨hal-05281501⟩
34 Consultations
0 Téléchargements

Partager

  • More