Convolutional neural network with parallel convolution scale attention module and ResCBAM for breast histology image classification.

Breast cancer is the most common cause of female morbidity and death worldwide. Compared with other cancers, early detection of breast cancer is more helpful to improve the prognosis of patients. In order to achieve early diagnosis and treatment, clinical treatment requires rapid and accurate diagnosis. Therefore, the development of an automatic detection system for breast cancer suitable for patient imaging is of great significance for assisting clinical treatment. Accurate classification of pathological images plays a key role in computer-aided medical diagnosis and prognosis. However, in the automatic recognition and classification methods of breast cancer pathological images, the scale information, the loss of image information caused by insufficient feature fusion, and the enormous structure of the model may lead to inaccurate or inefficient classification. To minimize the impact, we proposed a lightweight PCSAM-ResCBAM model based on two-stage convolutional neural network. The model included a Parallel Convolution Scale Attention Module network (PCSAM-Net) and a Residual Convolutional Block Attention Module network (ResCBAM-Net). The first-level convolutional network was built through a 4-layer PCSAM module to achieve prediction and classification of patches extracted from images. To optimize the network's ability to represent global features of images, we proposed a tiled feature fusion method to fuse patch features from the same image, and proposed a residual convolutional attention module. Based on the above, the second-level convolutional network was constructed to achieve predictive classification of images. We evaluated the performance of our proposed model on the ICIAR2018 dataset and the BreakHis dataset, respectively. Furthermore, through model ablation studies, we found that scale attention and dilated convolution play an important role in improving model performance. Our proposed model outperforms the existing state-of-the-art models on 200 × and 400 × magnification datasets with a maximum accuracy of 98.74 %.

miR-451a was selectively sorted into exosomes and promoted the progression of esophageal squamous cell carcinoma through CAB39.

Exosomes are emerging mediators of cell-cell communication, which are secreted from cells and may be delivered into recipient cells in cell biological processes. Here, we examined microRNA (miRNA) expression in esophageal squamous cell carcinoma (ESCC) cells. We performed miRNA sequencing in exosomes and cells of KYSE150 and KYSE450 cell lines. Among these differentially expressed miRNAs, 20 of the miRNAs were detected in cells and exosomes. A heat map indicated that the level of miR-451a was higher in exosomes than in ESCC cells. Furthermore, miRNA pull-down assays and combined exosomes proteomic data showed that miR-451a interacts with YWHAE. Over-expression of YWHAE leads to miR-451a accumulation in the exosomes instead of the donor cells. We found that miR-451a was sorted into exosomes. However, the biological function of miR-451a remains unclear in ESCC. Here, Dual-luciferase reporter assay was conducted and it was proved that CAB39 is a target gene of miR-451a. Moreover, CAB39 is related to TGF-ß1 from RNA-sequencing data of 155 paired of ESCC tissues and the matched tissues. Western Blot and qPCR revealed that CAB39 and TGF-ß1 were positively correlated in ESCC. Over-expression of CAB39 were cocultured with PBMCs from the blood from healthy donors. Flow cytometry assays showed that apoptotic cells were significantly reduced after CAB39 over-expression and significantly increased after treated with TGF-ß1 inhibitors. Thus, our data indicate that CAB39 weakens antitumor immunity through TGF-ß1 in ESCC. In summary, YWHAE selectively sorted miR-451a into exosomes and it can weaken antitumor immunity promotes tumor progression through CAB39.

Potential markers of cancer stem-like cells in ESCC: a review of the current knowledge.

In patients with esophageal squamous cell carcinoma (ESCC), the incidence and mortality rate of ESCC in our country are also higher than those in the rest of the world. Despite advances in the treatment department method, patient survival rates have not obviously improved, which often leads to treatment obstruction and cancer repeat. ESCC has special cells called cancer stem-like cells (CSLCs) with self-renewal and differentiation ability, which reflect the development process and prognosis of cancer. In this review, we evaluated CSLCs, which are identified from the expression of cell surface markers in ESCC. By inciting EMTs to participate in tumor migration and invasion, stem cells promote tumor redifferentiation. Some factors can inhibit the migration and invasion of ESCC via the EMT-related pathway. We here summarize the research progress on the surface markers of CSLCs, EMT pathway, and the microenvironment in the process of tumor growth. Thus, these data may be more valuable for clinical applications.

A Radiomics Nomogram for Non-Invasive Prediction of Progression-Free Survival in Esophageal Squamous Cell Carcinoma.

To construct a prognostic model for preoperative prediction on computed tomography (CT) images of esophageal squamous cell carcinoma (ESCC), we created radiomics signature with high throughput radiomics features extracted from CT images of 272 patients (204 in training and 68 in validation cohort). Multivariable logistic regression was applied to build the radiomics signature and the predictive nomogram model, which was composed of radiomics signature, traditional TNM stage, and clinical features. A total of 21 radiomics features were selected from 954 to build a radiomics signature which was significantly associated with progression-free survival (p < 0.001). The area under the curve of performance was 0.878 (95% CI: 0.831-0.924) for the training cohort and 0.857 (95% CI: 0.767-0.947) for the validation cohort. The radscore of signatures' combination showed significant discrimination for survival status. Radiomics nomogram combined radscore with TNM staging and showed considerable improvement over TNM staging alone in the training cohort (C-index, 0.770 vs. 0.603; p < 0.05), and it is the same with clinical data (C-index, 0.792 vs. 0.680; p < 0.05), which were confirmed in the validation cohort. Decision curve analysis showed that the model would receive a benefit when the threshold probability was between 0 and 0.9. Collectively, multiparametric CT-based radiomics nomograms provided improved prognostic ability in ESCC.

Survival prediction for patients with glioblastoma multiforme using a Cox proportional hazards denoising autoencoder network.

Objectives: This study aimed to establish and validate a prognostic model based on magnetic resonance imaging and clinical features to predict the survival time of patients with glioblastoma multiforme (GBM). Methods: In this study, a convolutional denoising autoencoder (DAE) network combined with the loss function of the Cox proportional hazard regression model was used to extract features for survival prediction. In addition, the Kaplan-Meier curve, the Schoenfeld residual analysis, the time-dependent receiver operating characteristic curve, the nomogram, and the calibration curve were performed to assess the survival prediction ability. Results: The concordance index (C-index) of the survival prediction model, which combines the DAE and the Cox proportional hazard regression model, reached 0.78 in the training set, 0.75 in the validation set, and 0.74 in the test set. Patients were divided into high- and low-risk groups based on the median prognostic index (PI). Kaplan-Meier curve was used for survival analysis (p = < 2e-16 in the training set, p = 3e-04 in the validation set, and p = 0.007 in the test set), which showed that the survival probability of different groups was significantly different, and the PI of the network played an influential role in the prediction of survival probability. In the residual verification of the PI, the fitting curve of the scatter plot was roughly parallel to the x-axis, and the p-value of the test was 0.11, proving that the PI and survival time were independent of each other and the survival prediction ability of the PI was less affected than survival time. The areas under the curve of the training set were 0.843, 0.871, 0.903, and 0.941; those of the validation set were 0.687, 0.895, 1.000, and 0.967; and those of the test set were 0.757, 0.852, 0.683, and 0.898. Conclusion: The survival prediction model, which combines the DAE and the Cox proportional hazard regression model, can effectively predict the prognosis of patients with GBM.

Comparative proteomic analysis of Aurelia coerulea for its locomotion system molecular structure-function inference.

BACKGROUND: Rhythmic contraction and autonomous movement play a key role in the predation, production and displacement of jellyfish. METHODS: Four independent body parts of the jellyfish Aurelia coerulea, including Bell, Tentacle, Oral arm and Gastric pouch were extracted and have been carried out a compared proteomics by liquid chromatography-mass spectrometry/mass-spectrometry (LC-MS/MS). ResultsA total of 13,429 peptides and 1916 proteins with molecular weights in the range of 10.6-980.9 kDa were identified, where 1916, 1562, 1474 and 1441 proteins were matched in the Gastric pouch, Tentacle, Oral arm and Bell, respectively. Gene Ontology (GO) analysis showed that translation, cytoplasma and ATP binding occupy the top differential terms of the three subdomains Biological process, Cellular Component and Molecular Function. A total of 326 pathways were successfully mapped that are mainly associated with intracellular synthesis, metabolism as well as intracellular functions. Moreover, a total of 27 contractile machinery associated proteins including 22 myosin, 3 actin and 2 tropomyosin were identified. CONCLUSIONS: Our results provide a composition profile in the four independent body parts of the jellyfish A. coerulea, of which the identified muscular proteins will greatly help in the understanding of the structural and functional relationship, as well as their operating mechanisms in the jellyfish locomotion system. SIGNIFICANCE: Omics studies have gained a new overall insight into the function of gene and protein networks during the development of motor systems in both bilateral and radial symmetrical animals. A compared proteomics using the label-free method of nano-LC-MS/MS has been performed through the four independent body parts of the moon jellyfish A. coerulea, including Bell, Tentacle, Oral arm and Gastric pouch. In addition to conventional bioinformatics analyses such as GO and KEGG, we have scanned the locomotion-related components, aligned their sequences, simulated three dimensional structures as well as did the molecular phylogenetic analyses. Our investigation provides a composition profile in the four independent body parts of the jellyfish A. coerulea, of which the identified muscular proteins will greatly help in the understanding of the structural and functional relationship, as well as their operating mechanisms in the jellyfish locomotion system.

De novo transcriptomic analysis of the venomous glands from the scorpion Heterometrus spinifer revealed unique and extremely high diversity of the venom peptides.

Scorpion, as an ancient species, has been widely used on dozens of human diseases in traditional Chinese Medicine. Although the scorpion venom from the Buthidae family with the potent toxicity attracts more interests, toxins from the non-Buthidae family draw great attention as well because of its abundance and complexity even without harm to mammals. Moreover, several toxic components of scorpion venom have been identified as valuable scaffolds for the drug design and development. Using the Next Generation Sequencing (NGS) technique, here we reported the transcriptome of the venomous glands of Heterometrus spinifer, a non-Buthidae scorpion that only a few toxic and complete components have been identified known-to-date. The total mRNA extracted from the venomous glands of H. spinifer was subjected to illumina sequencing with a strategy of de novo assembly, and a total of 54 189 transcripts were unigenes from a total of 88 311 600 determined reads. We annotated 18 567 (34.26%) unigenes from NR database, 12 258 (22.62%) from SWISSPROT database, 11 161 (20.60%) from GO database, 10 159 (18.75%) from COG database and 5059 (9.34%) from KEGG database, respectively. 2843 unigenes were further selected against the toxin-related sub-database of SWISSPROT. After removing the redundancy, 13 common toxin-related subfamilies with 62 unigenes were manually confirmed, including 8 K-toxins, 1 calcin, 3 Imperatoxin I-like, 2 La1-like, 1 scorpin-like, 3 antimicrobial peptides, two types of protease inhibitors such as 8 Kunitz-type protease inhibitors and 3 Ascaris-type protease inhibitors, and 33 proteases including 16 serine proteinases, 7 phospholipases, 5 metalloproteases, 3 hyaluronidases and 2 phosphatases. Our report is the first transcriptomic analyses of venomous glands from the scorpion H. spinifer, serving as a public information platform for the development of novel bio-therapeutics.