In clinical diagnosis of brain tumors, accurate segmentation based on multimodal magnetic resonance imaging (MRI) is essential for determining tumor type, extent, and spatial boundaries. However, differences in imaging mechanisms, information emphasis, and feature distributions among multimodal MRI data have posed significant challenges for precise tumor modeling and fusion-based segmentation. In recent years, fusion neural networks have provided effective strategies for integrating multimodal information and have become a major research focus in multimodal brain tumor segmentation. This review systematically summarized relevant studies on fusion neural networks for multimodal brain tumor segmentation published since 2019. First, the fundamental concepts of multimodal data fusion and model fusion were introduced. Then, existing methods were categorized into three types according to fusion levels: prediction fusion models, feature fusion models, and stage fusion models, and their structural characteristics and segmentation performance were comparatively analyzed. Finally, current limitations were discussed, and potential development trends of fusion neural networks for multimodal MRI brain tumor segmentation were summarized. This review aims to provide references for the design and optimization of future multimodal brain tumor segmentation models.
Objective To transfect bone marrow mesenchymal stem cells (BMSCs) of rats by recombinant adenovirus Ad-human matrix metalloproteinase 1 (hMMP-1) in vitro so as to lay the experimental foundation for the treatment of liver fibrosis with a combination of BMSCs and hMMP-1 gene transplantation. Methods BMSCs were isolated from bone marrow of 2-3 weeks old Sprague Dawley rats by whole bone marrow adherence method and identified, then transfected by recombinant adenovirus Ad-hMMP-1 carrying enhanced green fluorescent protein (EGFP) marker in vitro. The green fluorescent expression was observed by fluorescence microscope and the transfection efficiency was detected by flow cytometry to determine the optimum multiplicity of infection (MOI). BMSCs at passage 3 were divided into 3 groups: untransfected BMSCs group (group A), Ad-EGFP transfected BMSCs group (group B), and Ad-hMMP-1-EGFP transfected BMSCs group (group C); the gene and intracellular protein of hMMP-1 were detected by RT-PCR and Western blot; the ELISA assay was used to detect the supernatant protein expression, and the hMMP-1 activity was measured by fluorescent quantification kit. Results The green fluorescent was observed in BMSCs transfected by recombinant adenovirus at 24 hours after transfection; the fluorescence intensity was highest at 72 hours; and the optimum MOI was 200. The cells of 3 groups entered the logarithmic growth phase on the 3rd day and reached plateau phase on the 6th day by MTT assay; no significant difference was found in the cell proliferation rate among 3 groups (P gt; 0.05). RT-PCR, Western blot, and ELISA assay showed high expressions of the hMMP-1 gene and protein in group C, but no expression in groups A and B. The hMMP-1 activity was 1.24 nmol/(mg · min) in group C, but hMMP-1 activity was not detectable in groups A and B. Conclusion The exogenous hMMP-1 gene is successfully transfected into BMSCs of rats via recombinant adenovirus and can highly express, which lays the experimental foundation for the treatment of liver fibrosis with a combination of BMSCs and hMMP-1 gene transplantation.
Macrophages are important immune effector cells with significant plasticity and heterogeneity in the body immune system, and play an important role in normal physiological conditions and in the process of inflammation. It has been found that macrophage polarization involves a variety of cytokines and is a key link in immune regulation. Targeting macrophages by nanoparticles has a certain impact on the occurrence and development of a variety of diseases. Due to its characteristics, iron oxide nanoparticles have been used as the medium and carrier for cancer diagnosis and treatment, making full use of the special microenvironment of tumors to actively or passively aggregate drugs in tumor tissues, which has a good application prospect. However, the specific regulatory mechanism of reprogramming macrophages using iron oxide nanoparticles remains to be further explored. In this paper, the classification, polarization effect and metabolic mechanism of macrophages were firstly described. Secondly, the application of iron oxide nanoparticles and the induction of macrophage reprogramming were reviewed. Finally, the research prospect and difficulties and challenges of iron oxide nanoparticles were discussed to provide basic data and theoretical support for further research on the mechanism of the polarization effect of nanoparticles on macrophages.