Answers are grounded in this paper and cite their sources.
Seaweed is gaining attention for its potential as sustainable biomass. However, limited breeding efforts and research on crop management have been performed on seaweed compared to agricultural crops. Successful early development is critical for consistent yields, but high-throughput tools to study this microscopic life stage are lacking. Here we developed a high-throughput phenotyping pipeline for assessment of juvenile stages of kelp including semi-automated image acquisition and sporophyte trait by instance segmentation. We created a single large dataset comprising of 1129 images and 2823 instances which were reclassified to train four different models. Three sporophyte models were trained with different developmental stages to assess the effect of dataset composition on effectiveness and to provide flexibility for users with images of seaweed at different developmental stages. Our models achieve extremely high mean average precision scores (0.872 mAP50 and 0.681 mAP50-95) on mask detection, demonstrating excellent localization of juvenile sporophytes, even when overlapping at high sporophyte density. To follow growth and development or test treatment effects, the Alignment model was trained for alignment of images acquired sequentially over time, which achieved an mAP50 of 0.994 and mAP50-95 of 0.943 on masks. We compared the effectiveness and efficiency of the model with three experts. The model effectiveness was comparable to the experts, but 18x faster. This study provides an accurate, high-throughput phenotyping method for faster development of seaweed research that excludes inter-and intra-personal variability during data assessment. Potential applications are testing the effect of (a)biotic stresses, breeding programs and quality control of seeding material.
No takes yet. Share an insight, caveat, or question.
Put et al. (2026) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: