In computer vision applications, the primary task of object detection is to answer the following question: “What object is present and where is it located?”. However, underwater environments introduce challenges, such as poor lighting, high complexity, and diverse marine organism shapes, leading to missed detections or false positives in deep learning-based algorithms. To improve detection accuracy and robustness, this paper proposes an enhanced YOLOv11-based algorithm for underwater object detection that strengthens the ability to capture both local and global details and global contextual information in complex underwater environments. To better capture local and global features while integrating contextual information, the proposed method introduces several enhancements. The backbone incorporates a DualBottleneck module to enhance feature extraction, replacing the standard bottleneck structure in C3k, thus enhancing the feature extraction and the channel aggregation. The detection head adopts DyHead-GDC, integrating ghost depthwise separable convolution with DyHead for greater efficiency. Furthermore, the ADown module replaces conventional feature extraction and downsampling convolutions, reducing parameters and FLOPs by 14%. The C2PSF module, combining focal modulation and C2, strengthens local feature extraction and global context processing. Additionally, a SCSA module is inserted before the detection head to fully utilize multi-semantic information, improving the detection performance in complex underwater scenes. Experimental results confirm the effectiveness of these improvements. The model achieves 84.2% mAP50 on UTDAC2020, 84.4% on DUO and 86.7% on RUOD, surpassing the baseline by 2.5%, 1.6% and 1.2%, respectively. It remains lightweight, with 6.5 M parameters and a computational cost of 7.1 GFLOPs.

The Sichuan snub-nosed monkey (Rhinopithecus roxellana) is an endangered species in China and is susceptible to infection with Trichuris species. However, the worms are difficult to remove completely. A practical treatment regimen for trichuriasis was conducted over a seven-month period on 15 R. roxellana in a wildlife zoo in southern China. Initially, a combination of fecal examination, morphological observation, molecular identification of ITS1 and mitogenome, and infective pattern analysis revealed that R. roxellana was susceptible to Trichuris rhinopiptheroxella. Three rounds of treatment were administrated, with a dosage of 10 mg/kg albendazole or ivermectin in each. The initial administration of albendazole to 15 monkeys resulted in a reduction of fecal eggs per gram (EPG) by 7.8%–73.2%. The subsequent administration of albendazole to 9 monkeys demonstrated a reduction in fecal EPG by 52.6%–52.8%. The third administration of ivermectin to 5 monkeys resulted in a reduction of fecal EPG by 55.6%–96.6%. However, the EPG level increased in some monkeys after one month of these three anthelmintic administrations. Subsequently, improved strategies were implemented, including an increased dosage of albendazole, flame sterilization, the replacement of the sandy floor with concrete, and the hanging of food. Results showed that a reduction in the fecal EPG of 8 monkeys in the exhibition region, with a decrease from 5135 to 63. Additionally, 6 monkeys exhibited a negative EPG after one month. In the breeding region, the fecal EPG of 7 monkeys decreased from 7389 to 869. Additionally, 2 monkeys demonstrated a negative EPG after one month. This study provides evidence for the control of Trichuris infection in R. roxellana and offers a guideline for the treatment of trichuriasis in animals.

The Yangtze finless porpoises (Neophocaena asiaeorientalis asiaeorientalis) living in different environments display significant differences in behavior and physiology. To compare and analyze gene expression differences between an ex situ population and a controlled environment population of the Yangtze finless porpoise, we sequenced the transcriptome of blood tissues living in a semi-natural reserve and an artificial facility, respectively. We identified 6860 differentially expressed genes (DEGs), of which 6603 were up-regulated and 257 were down-regulated in the controlled environment vs ex situ comparison. GO and KEGG enrichment analysis showed that the up-regulated genes in the controlled environment population were significantly associated with glucose metabolism, amino acid metabolism, and the nervous system, while those up-regulated in the ex situ population were significantly associated with energy supply and biosynthesis. Further analysis showed that metabolic and hearing-related genes were significantly affected by changes in the environment, and key metabolic genes such as HK, PFK, IDH, and GLS and key hearing-related genes such as OTOA, OTOF, SLC38A1, and GABBR2 were identified. These results suggest that the controlled environment population may have enhanced glucose metabolic ability via activation of glycolysis/gluconeogenesis, the TCA cycle, and inositol phosphate metabolism, while the ex situ population may meet higher energy requirements via enhancement of the amino acid metabolism of the liver and muscle and oxidative phosphorylation. Additionally, the acoustic behavior and auditory-related genes of Yangtze finless porpoise may show responsive changes and differential expression under different environment conditions, and thus the auditory sensitivity may also show corresponding adaptive characteristics. This study provides a new perspective for further exploration of the responsive changes of the two populations to various environments and provides a theoretical reference for further improvements in conservation practices for the Yangtze finless porpoise.

Statistical analysis of the relationship between pre-parturition backfat thickness and reproductive performance using a large dataset of 693 sows (gilts) revealed that both high and low backfat as detrimental to the optimal reproductive performance of sows. Further analysis indicated that the health and reproductive performance of sows are compromised by problems such as excess backfat deposition due to ad libitum feeding and reduced satiety due to restricted feeding. Despite the consensus that dietary fiber (DF) supplementation of sows during gestation improves their health and performance, there is currently disagreement about the appropriate type of DF to select. In this review, the definition of DF is revised by integrating its physicochemical properties and fermentability, and the definition of functional dietary fiber (FDF) applicable to breeding pig breeding is proposed. Additionally, the mechanisms by which FDF regulates sow constipation, satiety, lactational feed intake and reproductive performance are elaborated in detail. This review provides a theoretical basis for selecting reasonable fiber sources for sow production.

The classification of high-frequency marine mammal vocalizations often faces challenges due to the limitations of acoustic features, which are sensitive to mid-to-low frequencies but offer low resolution in high-frequency ranges. Additionally, single-channel networks can restrict overall classification performance. To tackle these challenges, we introduce MT-Resformer, an innovative dual-channel model with a multi-scale framework designed for classifying marine mammal vocalizations. Our approach introduces a feature fusion strategy that combines the constant-Q spectrogram with Mel filter-based spectrogram features, effectively overcoming the low resolution of Mel spectrograms in high frequencies. The MT-Resformer model incorporates two key components: a multi-scale parallel residual network (MResNet) and a Transformer network channel. The model employs a multi-level neural perceptron (MLP) to dynamically regulate the weighting of the two channels, enabling flexible feature fusion. Experimental findings validate the proposed approach, yielding classification accuracies of 99.17% on the Watkins dataset and 95.22% on the ChangLong dataset. These results emphasize its outstanding performance.