Assessment of fish feeding intensity is crucial for achieving precise feeding and for enhancing aquaculture efficiency. However, complex backgrounds in real-world aquaculture scenarios—such as water surface reflections, wave disturbances, and the stochastic movement of fish schools—pose significant challenges to the precise extraction of feeding-related features. To address this issue, this study proposes a fish feeding intensity assessment method based on MobileViT-CoordAtt. The method employs a lightweight MobileViT backbone network, integrated with a Coordinate Attention (CoordAtt) mechanism and a multi-scale feature fusion strategy. Specifically, the CoordAtt module enhances the model’s spatial perception by encoding spatial coordinate information, enabling precise capture of the spatial distribution characteristics of fish schools. The multi-scale feature fusion strategy adopts a three-level feature integration approach (input features, local features, and global features) to further strengthen the model’s representational capacity, ensuring robust extraction of key feeding-related features across diverse scales and hierarchical levels. Experimental results demonstrate that the MobileViT-CoordAtt model, trained with transfer learning, achieves an accuracy of 97.18% on the test set, with a compact parameter size of 4.09 MB. These findings indicate that the proposed method can effectively evaluate fish feeding intensity in practical aquaculture environments, providing critical support for formulating dynamic feeding strategies.

In agricultural and fishery production, the occurrence of cyanobacterial blooms and the contamination of freshwater systems with microplastics have become increasingly important research focuses. However, the individual and combined toxic mechanisms of these two pollutants are not yet fully understood. Therefore, in this study, we analyzed the effects of these two stressors, microcystins (MC) and nanoplastics (NP), on the transcriptome and metabolome of the hepatopancreas of river clams. RNA and metabolites extracted from river clams treated with MC, NP, and a combination of MC and NP were used to construct standardized cDNA libraries, which were then subjected to integrated analysis. Significant enrichment of 49 pathways, 34 pathways, and 44 pathways was observed in the MC group compared to the control group, NP group compared to the control group, and NP-MC group compared to the control group, respectively. In these three experimental groups, we found that the lysosome pathway, which affects immune function and cell apoptosis, was enriched with numerous differentially expressed genes and metabolites. Changes in ATP6N and ADP may impair lysosomal acidification and disrupt normal lysosomal degradation processes, indicating interference with the hepatopancreatic metabolism of pollutants. Interestingly, we observed significant alterations in the cathepsin family, and the downregulation of cathepsin genes, along with the downregulation of ATP6N, implies a potential disruption in lysosomal proteolysis. In the NP-MC group, the downregulation of purine expression levels suggests an impact on the immune system of river clams by NP-MC. In conclusion, while there is some overlap in the damage caused to the hepatopancreas of river clams by MCs, MPs, and the combination of NP-MC, further research is necessary to fully understand their effects.

The presence of corticosteroids (CSs) in aquaculture wastewater poses risks to ecological health and food safety, yet data on their lipophilicity (logK_ow) remain scarce. This study determined the logK_ow of CSs to perform a screening-level in silico environmental risk prioritization. We evaluated nine computational programs (ACD/LogP, ALOGPS 2.1, CLOGP, JChem, KOWWIN, MiLogP, MolLogP, MOSES.logP, and XLOGP3) against experimental data for 50 steroid hormones. Results showed that XLOGP3 demonstrated the highest accuracy (Adjusted R² = 0.9872; SSE = 0.1004), followed by MiLogP, ACD/LogP, and KOWWIN. Structure–lipophilicity analysis revealed that esterification and acetonide formation significantly increase logK_ow, while hydroxylation decreases it. Using the validated XLOGP3, we predicted logK_ow for 32 synthetic CSs and estimated their bioconcentration factor (BCF) and soil organic carbon–water partition coefficient (K_oc). Because experimental logK_ow data for these 32 synthetic compounds are largely unavailable, these estimates should be interpreted as preliminary prioritization indicators rather than experimentally confirmed endpoints. Heavily modified CSs like Ciclesonide and Fluocortolone 21-hexanoate exhibited high logK_ow (>4.5), log BCF (>3.0), and logK_oc (>4.0), indicating their high potential for bioaccumulation and persistent sediment adsorption. This study provides a prioritized list of high-risk CSs, serving as a preliminary tool to identify potential compounds of concern in aquaculture environments.