Accurate body dimension and weight measurements are critical for optimizing poultry management, health assessment, and economic efficiency. This study introduces an innovative deep learning-based model leveraging multimodal data—2D RGB images from different views, depth images, and 3D point clouds—for the non-invasive estimation of duck body dimensions and weight. A dataset of 1023 Linwu ducks, comprising over 5000 samples with diverse postures and conditions, was collected to support model training. The proposed method innovatively employs PointNet++ to extract key feature points from point clouds, extracts and computes corresponding 3D geometric features, and fuses them with multi-view convolutional 2D features. A Transformer encoder is then utilized to capture long-range dependencies and refine feature interactions, thereby enhancing prediction robustness. The model achieved a mean absolute percentage error (MAPE) of 5.73% and an R² of 0.953 across seven morphometric parameters describing body dimensions, and an MAPE of 10.49% with an R² of 0.952 for body weight, indicating robust and consistent predictive performance across both structural and mass-related phenotypes. Unlike conventional manual measurements, the proposed model enables high-precision estimation while eliminating the necessity for physical handling, thereby reducing animal stress and broadening its application scope. This study marks the first application of deep learning techniques to poultry body dimension and weight estimation, providing a valuable reference for the intelligent and precise management of the livestock industry with far-reaching practical significance.

Dietary flavonoids are increasingly investigated in swine nutrition, but how supplementation level, intervention duration, and growth stage are associated with growth and antioxidant responses remains unclear. This PRISMA 2020 systematic review and meta-analysis evaluated dietary flavonoid supplementation in pigs. Twenty-eight randomized controlled trials were synthesized using three-level random-effects models to account for dependent effect sizes. Flavonoid supplementation was associated with increased average daily gain (ADG; SMD = 0.719, 95% CI: 0.428 to 1.009; $I^2=51.85\%$) and a lower feed-to-gain ratio (F/G; SMD = −0.708, 95% CI: −0.929 to −0.487; $I^2=32.25\%$), whereas the association with average daily feed intake (ADFI) was not statistically significant (SMD = 0.217, 95% CI: −0.101 to 0.535; $I^2=59.30\%$). Supplementation was also associated with increased superoxide dismutase (SMD = 0.907) and glutathione peroxidase activities (SMD = 0.666) and reduced malondialdehyde concentrations (SMD = −0.989), although heterogeneity was substantial for these outcomes ($I^2=96.47$–$97.83\%$). The certainty of evidence was assessed as low for ADG and moderate for ADFI, F/G, and antioxidant outcomes under the adapted GRADE framework. Exploratory stage-specific dose–time meta-regression identified potential relationships between supplementation intensity and ADG in weaned pigs (35 effect sizes from 14 studies) and finishing pigs (12 effect sizes from four studies). In finishing pigs, intervention duration was associated with growth responses. However, this subgroup contained few independent studies, and intervention duration covaried with age, physiological stage, flavonoid source, dose, health status, and basal diet. This association therefore does not establish an independent causal effect of intervention duration.