Papers

DFMA: an improved DeepLabv3+ based on FasterNet, multi-receptive field, and attention mechanism for high-throughput phenotyping of seedlings

Researchers developed an improved deep learning model called DFMA for automated measurement of plant seedling length, a key metric for assessing seed viability. The model achieved high accuracy on rice seedling and other plant datasets, outperforming existing approaches in generating detailed segmentation masks of seedling structures. While not directly about microplastics, the technology addresses agricultural phenotyping challenges relevant to understanding crop responses to environmental stressors.

Advancement and Challenges of Microplastic Pollution in the Aquatic Environment: a Review

Synthetic microbiota for microplastic degradation modulates rhizosphere fungal diversity and metabolic function in highland barley

Researchers examined how a synthetic microbiota consortium (MPDSM) designed for microplastic degradation affects rhizosphere fungal diversity and nutritional quality in highland barley grown in polystyrene-contaminated soil. The MPDSM achieved up to 19.9% weight loss in large microplastic particles and significantly modulated rhizosphere fungal metabolic function, suggesting microbiome-based remediation can partly offset crop quality impacts.