Automatically segmenting polyps from colonoscopy videos is crucial for developing computer-assisted diagnostic systems for colorectal cancer. Existing automatic polyp segmentation methods often struggle to fulfill the real-time demands of clinical applications due to their substantial parameter count and computational load, especially those based on Transformer architectures. To tackle these challenges, a novel lightweight long-range context fusion network, named LightCF-Net, is proposed in this paper. This network attempts to model long-range spatial dependencies while maintaining real-time performance, to better distinguish polyps from background noise and thus improve segmentation accuracy. A novel Fusion Attention Encoder (FAEncoder) is designed in the proposed network, which integrates Large Kernel Attention (LKA) and channel attention mechanisms to extract deep representational features of polyps and unearth long-range dependencies. Furthermore, a newly designed Visual Attention Mamba module (VAM) is added to the skip connections, modeling long-range context dependencies in the encoder-extracted features and reducing background noise interference through the attention mechanism. Finally, a Pyramid Split Attention module (PSA) is used in the bottleneck layer to extract richer multi-scale contextual features. The proposed method was thoroughly evaluated on four renowned polyp segmentation datasets: Kvasir-SEG, CVC-ClinicDB, BKAI-IGH, and ETIS. Experimental findings demonstrate that the proposed method delivers higher segmentation accuracy in less time, consistently outperforming the most advanced lightweight polyp segmentation networks.

The increasing global demand for skin disease diagnostics emphasizes the urgent need for advancements in AI-assisted diagnostic technologies for dermatoscopic images. In current practical medical systems, the primary challenge is balancing lightweight models with accurate image analysis to address constraints like limited storage and computational costs. While knowledge distillation methods hold immense potential in healthcare applications, related research on multi-class skin disease tasks is scarce. To bridge this gap, our study introduces an enhanced multi-source knowledge fusion distillation framework, termed DSP-KD, which improves knowledge transfer in a dual-stage progressive distillation approach to maximize mutual information between teacher and student representations. The experimental results highlight the superior performance of our distilled ShuffleNetV2 on both the ISIC2019 dataset and our private skin disorders dataset. Compared to other state-of-the-art distillation methods using diverse knowledge sources, the DSP-KD demonstrates remarkable effectiveness with a smaller computational burden.

Introduction: GBS (group B streptococcus) is an opportunistic pathogen that can colonize healthy individuals but presents significant challenges in clinical obstetrics and gynecology, as it can cause miscarriage, preterm birth, and invasive infections in newborns. To develop specific and personalized preventative strategies, a better understanding of the epidemiological characteristics and pathogenic features of GBS is essential.Methods: We conducted a comprehensive strain-level genomic analysis of GBS, examining serotype and genotype distributions, as well as the composition and correlations of virulence genes using the blastn-short mode of the BLAST program(v2.10.0+), mlstsoftware (https://github.com/tseemann/mlst), Snippy (v4.6.0), FastTree (v2.1.11) and iTOL. The coding sequence region of virulence factors was annotated by Prodigal (v2.6.3) and Glimmer(v3.02b). We further identified host protein interacting with Srr2 by mass spectrometry analysis.Results: While certain genotypes showed strong serotype consistency, there was no significant association between overall serotypes and genotypes. However, the composition of virulence genes was more closely related to the phylogeny of GBS, among which simultaneous presence of Srr2 and HygA exhibit significant association with hypervirulence. Tubulin emerged as the most distinct and abundant hit. The specific interaction of Tubulin with Srr2-BR, rather than Srr1-BR, was further confirmed by immunoblotting.Discussion: Considering the impact of cytoskeleton rearrangement on GBS pathogenesis, this observation offers a plausible explanation for the hypervirulence triggered by Srr2. Collectively, our findings indicate that in the future clinical practice, virulence gene detection should be given more attention to achieve precise GBS surveillance and disease prevention.

Rapid identification of Candida species is significant for the diagnosis of vulvovaginal candidiasis (VVC). An integrated and multi-target system for the rapid, high-specificity, and high-sensitivity detection of four Candida species was developed. The system consists of a rapid sample processing cassette and a rapid nucleic acid analysis device. The cassette could process the Candida species to release nucleic acids in 15 min. The released nucleic acids were analyzed by the device as fast as within 30 min, using the loop-mediated isothermal amplification method. The four Candida species could be simultaneously identified, with each reaction using only 1.41 µL of reaction mixture, which was low cost. The RPT (rapid sample processing and testing) system could detect the four Candida species with high sensitivity (90 days, and the system could also be used for the detection of bacteria.

A microfluidic-chip system is developed using LAMP for fast and simultaneous detection of Trichomonas vaginalis and high-risk human papillomavirus, and it is promisingly applicable in various fields, particularly in co-infection diagnosis.

With the further development of neural networks, automatic segmentation techniques for melanoma are becoming increasingly mature, especially under the conditions of abundant hardware resources. This allows for the accuracy of segmentation to be improved by increasing the complexity and computational capacity of the model. However, a new problem arises when it comes to actual applications, as there may not be the high-end hardware available, especially in hospitals and among the general public, who may have limited computing resources. In response to this situation, this paper proposes a lightweight deep learning network that can achieve high segmentation accuracy with minimal resource consumption. We introduce a network called DTONet (double-tailed octave network), which was specifically designed for this purpose. Its computational parameter count is only 30,859, which is 1/256th of the mainstream UNet model. Despite its reduced complexity, DTONet demonstrates superior performance in terms of accuracy, with an IOU improvement over other similar models. To validate the generalization capability of this model, we conducted tests on the PH2 dataset, and the results still outperformed existing models. Therefore, the proposed DTONet network exhibits excellent generalization ability and is sufficiently outstanding.