Medical cross-modal retrieval aims to achieve semantic similarity search between different modalities of medical cases, such as quickly locating relevant ultrasound images through ultrasound reports, or using ultrasound images to retrieve matching reports. However, existing medical cross-modal hash retrieval methods face significant challenges, including semantic and visual differences between modalities and the scalability issues of hash algorithms in handling large-scale data. To address these challenges, this paper proposes a Medical image Semantic Alignment Cross-modal Hashing based on Transformer (MSACH). The algorithm employed a segmented training strategy, combining modality feature extraction and hash function learning, effectively extracting low-dimensional features containing important semantic information. A Transformer encoder was used for cross-modal semantic learning. By introducing manifold similarity constraints, balance constraints, and a linear classification network constraint, the algorithm enhanced the discriminability of the hash codes. Experimental results demonstrated that the MSACH algorithm improved the mean average precision (MAP) by 11.8% and 12.8% on two datasets compared to traditional methods. The algorithm exhibits outstanding performance in enhancing retrieval accuracy and handling large-scale medical data, showing promising potential for practical applications.
In order to address the issues of spatial induction bias and lack of effective representation of global contextual information in colon polyp image segmentation, which lead to the loss of edge details and mis-segmentation of lesion areas, a colon polyp segmentation method that combines Transformer and cross-level phase-awareness is proposed. The method started from the perspective of global feature transformation, and used a hierarchical Transformer encoder to extract semantic information and spatial details of lesion areas layer by layer. Secondly, a phase-aware fusion module (PAFM) was designed to capture cross-level interaction information and effectively aggregate multi-scale contextual information. Thirdly, a position oriented functional module (POF) was designed to effectively integrate global and local feature information, fill in semantic gaps, and suppress background noise. Fourthly, a residual axis reverse attention module (RA-IA) was used to improve the network’s ability to recognize edge pixels. The proposed method was experimentally tested on public datasets CVC-ClinicDB, Kvasir, CVC-ColonDB, and EITS, with Dice similarity coefficients of 94.04%, 92.04%, 80.78%, and 76.80%, respectively, and mean intersection over union of 89.31%, 86.81%, 73.55%, and 69.10%, respectively. The simulation experimental results show that the proposed method can effectively segment colon polyp images, providing a new window for the diagnosis of colon polyps.
Leukemia is a common, multiple and dangerous blood disease, whose early diagnosis and treatment are very important. At present, the diagnosis of leukemia heavily relies on morphological examination of blood cell images by pathologists, which is tedious and time-consuming. Meanwhile, the diagnostic results are highly subjective, which may lead to misdiagnosis and missed diagnosis. To address the gap above, we proposed an improved Vision Transformer model for blood cell recognition. First, a faster R-CNN network was used to locate and extract individual blood cell slices from original images. Then, we split the single-cell image into multiple image patches and put them into the encoder layer for feature extraction. Based on the self-attention mechanism of the Transformer, we proposed a sparse attention module which could focus on the discriminative parts of blood cell images and improve the fine-grained feature representation ability of the model. Finally, a contrastive loss function was adopted to further increase the inter-class difference and intra-class consistency of the extracted features. Experimental results showed that the proposed module outperformed the other approaches and significantly improved the accuracy to 91.96% on the Munich single-cell morphological dataset of leukocytes, which is expected to provide a reference for physicians’ clinical diagnosis.
Manual segmentation of coronary arteries in computed tomography angiography (CTA) images is inefficient, and existing deep learning segmentation models often exhibit low accuracy on coronary artery images. Inspired by the Transformer architecture, this paper proposes a novel segmentation model, the double parallel encoder u-net with transformers (DUNETR). This network employed a dual-encoder design integrating Transformers and convolutional neural networks (CNNs). The Transformer encoder transformed three-dimensional (3D) coronary artery data into a one-dimensional (1D) sequential problem, effectively capturing global multi-scale feature information. Meanwhile, the CNN encoder extracted local features of the 3D coronary arteries. The complementary features extracted by the two encoders were fused through the noise reduction feature fusion (NRFF) module and passed to the decoder. Experimental results on a public dataset demonstrated that the proposed DUNETR model achieved a Dice similarity coefficient of 81.19% and a recall rate of 80.18%, representing improvements of 0.49% and 0.46%, respectively, over the next best model in comparative experiments. These results surpassed those of other conventional deep learning methods. The integration of Transformers and CNNs as dual encoders enables the extraction of rich feature information, significantly enhancing the effectiveness of 3D coronary artery segmentation. Additionally, this model provides a novel approach for segmenting other vascular structures.
The synergistic effect of drug combinations can solve the problem of acquired resistance to single drug therapy and has great potential for the treatment of complex diseases such as cancer. In this study, to explore the impact of interactions between different drug molecules on the effect of anticancer drugs, we proposed a Transformer-based deep learning prediction model—SMILESynergy. First, the drug text data—simplified molecular input line entry system (SMILES) were used to represent the drug molecules, and drug molecule isomers were generated through SMILES Enumeration for data augmentation. Then, the attention mechanism in the Transformer was used to encode and decode the drug molecules after data augmentation, and finally, a multi-layer perceptron (MLP) was connected to obtain the synergy value of the drugs. Experimental results showed that our model had a mean squared error of 51.34 in regression analysis, an accuracy of 0.97 in classification analysis, and better predictive performance than the DeepSynergy and MulinputSynergy models. SMILESynergy offers improved predictive performance to assist researchers in rapidly screening optimal drug combinations to improve cancer treatment outcomes.
Objective To observe the differences in protein contents of three transforming growth factorbeta(TGF-β) isoforms, β1, β2, β3 andtheir receptor(I) in hypertrophic scar and normal skin and to explore their influence on scar formation. Methods Eight cases of hypertrophic scar and their corresponding normal skin were detected to compare the expression and distribution of TGF-β1, β2, β3 and receptor(I) with immunohistochemistry and common pathological methods. Results Positive signals of TGF-β1, β2, and β3 could all be deteted in normal skin, mainly in the cytoplasm and extracellular matrix of epidermal cells; in addition, those factors could also be found in interfollicular keratinocytes and sweat gland cells; and the positive particles of TGF-β R(I) were mostly located in the membrane of keratinocytes and some fibroblasts. In hypertrophic scar, TGF-β1 and β3 could be detected in epidermal basal cells; TGFβ2 chiefly distributed in epidermal cells and some fibroblast cells; the protein contents of TGF-β1 and β3 were significantly lower than that of normal skin, while the change of TGF-β2 content was undistinguished when compared withnormalskin. In two kinds of tissues, the distribution and the content of TGF-β R(I) hadno obviously difference. ConclusionThe different expression and distribution of TGF-β1, β2 andβ3 between hypertrophic scar and normal skin may beassociated with the mechanism controlling scar formation, in which the role of the TGF-βR (I) and downstream signal factors need to be further studied.
ObjectiveTo study the effect of transforming growth factor β3 (TGF-β3), bone morphogenetic protein 2 (BMP-2), and dexamethasone (DEX) on the chondrogenic differentiation of rabbit synovial mesenchymal stem cells (SMSCs). MethodsSMSCs were isolated from the knee joints of 5 rabbits (weighing, 1.8-2.5 kg), and were identified by morphogenetic observation, flow cytometry detection for cell surface antigen, and adipogenic and osteogenic differentiations. The SMSCs were cultured in the PELLET system for chondrogenic differentiation. The cell pellets were divided into 8 groups: TGF-β3 was added in group A, BMP-2 in group B, DEX in group C, TGF-β3+BMP-2 in group C, TGF-β3+DEX in group E, BMP-2+DEX in group F, and TGF-β3+BMP-2+DEX in group G; group H served as control group. The diameter, weight, collagen type II (immuohistochemistry staining), proteoglycan (toluidine blue staining), and expression of cartilage related genes [real time quantitative PCR (RT-qPCR) technique] were compared to evaluate the effect of cytokines on the chondrogenic differentiation of SMSCs. Meanwhile, the DNA content of cell pellets was tested to assess the relationship between the increase weight of cell pellets and the cell proliferation. ResultsSMSCs were isolated from the knee joints of rabbits successfully and the findings indicated that the rabbit synovium-derived cells had characteristics of mesenchymal stem cells. The diameter, weight, collagen type II, proteoglycan, and expression of cartilage related genes of pellets in groups A-F were significantly lower than those of group G (P<0.05). RT-qPCR detection results showed that the relative expressions of cartilage related genes (SOX-9, Aggrecan, collagen type II, collagen type X, and BMP receptor II) in group G were significantly higher than those in the other groups (P<0.01). Meanwhile, with the increase of the volume of pellet, the DNA content reduced about 70% at 7 days, about 80% at 14 days, and about 88% at 21 days. ConclusionThe combination of TGF-β3, BMP-2, and DEX can make the capacity of chondrogenesis of SMSCs maximized. The increase of the pellet volume is caused by the extracellular matrix rather than by cell proliferation.
OBJECTIVE: To localize the distribution of basic fibroblast growth factor (bFGF) and transforming growth factor-beta(TGF-beta) in tissues from dermal chronic ulcer and hypertrophic scar and to explore their effects on tissue repair. METHODS: Twenty-one cases were detected to localize the distribution of bFGF and TGF-beta, among them, there were 8 cases with dermal chronic ulcers, 8 cases with hypertrophic scars, and 5 cases of normal skin. RESULTS: Positive signal of bFGF and TGF-beta could be found in normal skin, mainly in the keratinocytes. In dermal chronic ulcers, positive signal of bFGF and TGF-beta could be found in granulation tissues. bFGF was localized mainly in fibroblasts cells and endothelial cells and TGF-beta mainly in inflammatory cells. In hypertrophic scar, the localization and signal density of bFGF was similar with those in granulation tissues, but the staining of TGF-beta was negative. CONCLUSION: The different distribution of bFGF and TGF-beta in dermal chronic ulcer and hypertrophic scar may be the reason of different results of tissue repair. The pathogenesis of wound healing delay in a condition of high concentration of growth factors may come from the binding disorder of growth factors and their receptors. bFGF may be involved in all process of formation of hypertrophic scar, but TGF-beta may only play roles in the early stage.
OBJECTIVE This paper was to study the biological characteristics of the transformed human embryonic tendon cells, the relation between cell growth and culture conditions, and to compare these features with that of human embryonic tendon cells. METHODS The pts A58H plasmid had successfully used to transform a tendon cell line from human embryo in our past work. The human embryonic tendon cells and the transformed human embryonic tendon cells were cultured in vitro. In different culture conditions, the growth curve were drawn respectively. Population dependence and proliferation capability of the cells were investigated through plate cloning test and soft agar culture. The collagen secreted by cells was identified by immunohistochemical method. RESULTS In routine culture condition, the growth properties of the human embryonic tendon cell and transformed cells were almost identical. The growth properties of the transformed cells were not changed when the cells were frozen storage. There were changes of growth characteristics of the transformed cells when the culture temperature was changed. The transformed cells could subcultured continually and permanently. The proliferation capability of the transformed cells were ber than that of the human embryonic tendon cells. Moreover, the growth of the transformed cells was serum-dependent, and the phenomenon of contact inhibition was observed. The transformed cells were not able to grow on soft agar culture. They had the capacity of secreting collagen type I. CONCLUSION The transformed human embryonic tendon cells could be subcultured continually and permanently, and their growth could be controlled by changing their culture conditions and they had no malignant tendency in biological characteristics. They could be taken as an ideal experimental material for tendon engineering.
Objective To study biological effect of transforming growth factor β(TGF-β) and recombinant human bone morphogenetic protein 2 (rhBMP-2) on theSchwann cell(SC) in vitro. Methods Cultured SC from newbornSDrats were implanted at 5×103/well in 96-well-plate (36 wells in each group, altogether 3 groups):TGF-β group (group A) treated with 50 ng/ml TGF-β; rhBMP-2 group (group B) treated with 50 ng/ml rhBMP-2 and control group (group C). SC proliferation activity was assessed by MTT and flow cytometry (FCM) methods, and nerve growth factor (NGF) synthesis in SC culture media was detected by ELISA method. Results MTT observation indicated that there was significant difference in the growth curve among 3 groups until the 8th and 9th day. Group A had more obvious rising tendency than group B and group C. FCM observation indicated that the proliferation index of group A and group B was higher than that of group C（Plt;0.05）. ELISA observation indicated that there was significant difference in the NGF concentration of the culture medium among the 3 groups(P＜0.05). Group A had the highest NGF concentration. Conclusion Exogenous TGF-β and rhBMP-2 can promote SC’s ability to proliferate NGF, but TGF-β is more effective than rhBMP-2.