Background: Medical image segmentation is an important processing step in most of medical image analysis. Thus, high accuracy and robustness are required for them. The current deep neural network based medical segmentation methods have good effect on image with balanced foreground and background, but it will loss the characteristics of small targets on image with imbalanced foreground and background after multiple convolutions. Methods: In order to retain the features of small targets in the deep network, we proposed a new medical image segmentation model based on the U-Net with squeeze-and-excitation and attention modules which form a spiral closed path,callled as Spiral Squeeze-and-Excitation and Attention NET (SEA-NET) in this paper. The segmentation model used squeeze-and-extraction modules to adjust the channel information to enhance the useful information and used attention modules to adjust the spatial information of the feature map to highlight the target area for small target segmentation when up-sampling. The deep semantic information is integrated into the shallow feature map by the attention model. Therefore, the deep semantic information cannot be scattered by continuous up-sampling. We used cross entropy loss + Tversky loss function for fast convergence and well processing the imbalanced data sets. Our proposed SEA-NET was tested on the brain MRI dataset LPBA40 and peripheral blood smear images. Conclusions: On brain MRI data, the average value of the Dice coefficient we obtained reached 98.1 $$\%$$% . On the peripheral blood smear dataset, our proposed model has a good segmentation effect on adhesion cells. Results: The experimental results proved that the proposed SEA-Net performed better than U-Net, U-Net++, etc. in medical image segmentation.

Mitochondrial permeability transition (MPT)-driven necrosis (MPTDN) was a regulated variant of cell death triggered by specific stimuli. It played a crucial role in the development of organisms and the pathogenesis of diseases, and may provide new strategies for treating various diseases. However, there was limited research on the mechanisms of MPTDN in cervical cancer (CESC) at present. In this study, Weighted Gene Co-expression Network Analysis (WGCNA) was performed on differentially expressed genes in CESC. The module MEyellow, which showed the highest correlation with the phenotype, was selected for in-depth analysis. It was found that the genes in the MEyellow module may be associated with the tumor immune microenvironment (TIME). Through COX univariate regression and LASSO regression analysis, 6 key genes were identified. These genes were further investigated from multiple perspectives, including their independent diagnostic value, prognostic value, specific regulatory mechanisms in the tumor immune microenvironment, drug sensitivity analysis, and somatic mutation analysis. This study provided a comprehensive exploration of the mechanisms of action of these 6 key genes in CESC patients. And qRT-PCR validation was also conducted. Through COX univariate regression and LASSO coefficient screening of the MEyellow module, 6 key genes were identified: CHRM3-AS2, AC096734.1, BISPR, LINC02446, LINC00944, and DGUOK-AS1. Evaluation of the independent diagnostic value of these 6 key genes revealed that they can serve as independent diagnostic biomarkers. Through correlation analysis among these 6 genes, a potential regulatory mechanism among them was identified. Therefore, a risk prognostic model was established based on the collective action of these 6 genes, and the model showed good performance in predicting the survival period of CESC patients. By studying the relationship between these 6 key genes and the tumor microenvironment of CESC patients from multiple angles, it was found that these 6 genes are key regulatory factors in the tumor immune microenvironment of CESC patients. Additionally, 16 drugs that are associated with these 6 key genes were identified, and 8 small molecule drugs were predicted based on the lncRNA-mRNA network. The 6 key genes can serve as independent biomarkers for diagnosis, and the Risk score of these genes when acting together can be used as an indicator for predicting the clinical survival period of CESC patients. Additionally, these 6 key genes were closely related to the tumor immune microenvironment of CESC patients and were the important regulatory factors in the tumor immune microenvironment of CESC patients.

Background: Infective Endocarditis (IE) and Sepsis are two closely related infectious diseases, yet their shared pathogenic mechanisms at the transcriptional level remain unclear. This research gap poses a barrier to the development of refined therapeutic strategies and drug innovation.Methods: This study employed a collaborative approach using both microarray data and single-cell RNA sequencing (scRNA-seq) data to identify biomarkers for IE and Sepsis. It also offered an in-depth analysis of the roles and regulatory patterns of immune cells in these diseases.Results: We successfully identified four key biomarkers correlated with IE and Sepsis, namely CD177, IRAK3, RNASE2, and S100A12. Further investigation revealed the central role of Th1 cells, B cells, T cells, and IL-10, among other immune cells and cytokines, in the pathogenesis of these conditions. Notably, the small molecule drug Matrine exhibited potential therapeutic effects by targeting IL-10. Additionally, we discovered two Sepsis subgroups with distinct inflammatory responses and therapeutic strategies, where CD177 demonstrated significant classification value. The reliability of CD177 as a biomarker was further validated through qRT-PCR experiments.Conclusion: This research not only paves the way for early diagnosis and treatment of IE and Sepsis but also underscores the importance of identifying shared pathogenic mechanisms and novel therapeutic targets at the transcriptional level. Despite limitations in data volume and experimental validation, these preliminary findings add new perspectives to our understanding of these complex diseases.

Brain extraction is an essential pre-processing step for neuroimaging analysis. It is difficult to achieve high-precision extraction from low-quality brain MRI images with artifacts and gray inconsistencies which often result in irregular hole regions in the extracted brain tissues. In addition, the U-Net based brain extraction method trends to output over-smoothed brain boundary. To remove those irregular holes in the extracted mask, we proposed a new U-Net based model for brain extraction named O-Net. O-Net replaces the skip-connection path in the U-Net with dual shortcut paths including an attention module of an O-shaped network, which uses deep semantic information to highlight the target area while retaining more image details. O-Net effectively reduces the impact of intensity differences caused by artifacts or gray inconsistencies in the brain MRI images on the extraction results. To more accurately identify brain boundary, we designed a new GAN based brain extraction method, which used above O-Net as the segmentation network. The discrimination network of the proposed GAN model adopts the residual structure to enhance the nonlinear expression ability of the network to balance the adversarial training of the two networks. To speed up the convergence of the proposed model, a segmentation loss was added to the adversarial loss to supervise the feature learning of the segmentation network. This method was compared with other popular brain extraction methods on two public datasets (IBSR18 and LPBA40). The mean Dice similarity coefficients obtained by the proposed method were 97.26% and 98.29% on IBSR18 and LPBA40 respectively. In the comparative experiment, the results of the proposed method are the best on the two public datasets. Experimental results show that the proposed model can stably output high-precision brain tissue extraction images and the influence of artifacts and gray inconsistencies is small.

The extraction of brain tissue from cerebral MRI volume is an important pre-procedure for neuroimage analyses. The authors have developed an accurate and robust brain extraction method using a hybrid level set based active contour neighborhood model. The method uses a nonlinear speed function in the hybrid level set model to eliminate boundary leakage. When using the new hybrid level set model an active contour neighborhood model is applied iteratively in the neighborhood of brain boundary. A slice by slice contour initial method is proposed to obtain the neighborhood of the brain boundary. The method was applied to the internet brain MRI data provided by the Internet Brain Segmentation Repository (IBSR). In testing, a mean Dice similarity coefficient of 0.95±0.02 and a mean Hausdorff distance of 12.4±4.5 were obtained when performing our method across the IBSR data set (18 × 1.5 mm scans). The results obtained using our method were very similar to those produced using manual segmentation and achieved the smallest mean Hausdorff distance on the IBSR data. An automatic method of brain extraction from cerebral MRI volume was achieved and produced competitively accurate results.