Automated phenotyping of mammary gland tissues using computer vision systems.

Quantifying mammary gland tissue composition is crucial for understanding growth patterns and future lactation potential in dairy heifers, yet conventional histological analysis is labor-intensive and time-consuming. This study developed an automated phenotyping framework using deep learning-based semantic segmentation to classify tissues in mammary gland biopsies collected from Holstein heifers at 10, 26, 39, and 52 wk of age. A dataset of 130 histological images was manually annotated and divided into 383,650 patches of 448 × 448 pixels to accommodate the architecture requirements of pretrained models. U-Net and UNet++ models with ResNet34 encoders were trained using focal loss and evaluated through intersection over union (IoU) metrics. The best-performing model was a UNet++ without attention module, achieving high IoU values across validation and test sets. Mean average percentage error remained below 5% for most tissue-time combinations, with ducts demonstrating the highest prediction accuracy despite their limited presence in early developmental stages. Adipose tissue showed the most dynamic growth pattern, increasing from ∼2.5% of whole tissue at 10 wk of age to more than 30% at 26 and 39 wk of age before declining to less than 25% at 52 wk of age. Strong correlations between predicted and observed pixel counts confirmed model reliability across tissues and time points. This automated approach provides a scalable, reproducible solution for histological phenotyping, enabling high-throughput quantification of mammary tissue development that was previously impractical using manual methods. The framework successfully bridges the gap between detailed tissue characterization and the demands of large-scale phenotyping studies in dairy science.