Application of the convolution neural network in determining the depth of invasion of gastrointestinal cancer: a systematic review and meta-analysis.

BACKGROUND: With the development of endoscopic technology, endoscopic submucosal dissection (ESD) has been widely used in the treatment of gastrointestinal tumors. It is necessary to evaluate the depth of tumor invasion before the application of ESD. The convolution neural network (CNN) is a type of artificial intelligence that has the potential to assist in the classification of the depth of invasion in endoscopic images. This meta-analysis aimed to evaluate the performance of CNN in determining the depth of invasion of gastrointestinal tumors. METHODS: A search on PubMed, Web of Science, and SinoMed was performed to collect the original publications about the use of CNN in determining the depth of invasion of gastrointestinal neoplasms. Pooled sensitivity and specificity were calculated using an exact binominal rendition of the bivariate mixed-effects regression model. I2 was used for the evaluation of heterogeneity. RESULTS: A total of 17 articles were included; the pooled sensitivity was 84% (95% CI, 0.81-0.88), specificity was 91% (95% CI, 0.85-0.94), and the area under the curve (AUC) was 0.93 (95% CI, 0.90-0.95). The performance of CNN was significantly better than that of endoscopists (AUC: 0.93 vs 0.83, respectively; P = .0005). CONCLUSION: Our review revealed that CNN is one of the most effective methods of endoscopy to evaluate the depth of invasion of early gastrointestinal tumors, which has the potential to work as a remarkable tool for clinical endoscopists to make decisions on whether the lesion is feasible for endoscopic treatment.

A Deep Learning Application of Capsule Endoscopic Gastric Structure Recognition Based on a Transformer Model.

BACKGROUND: Gastric structure recognition systems have become increasingly necessary for the accurate diagnosis of gastric lesions in capsule endoscopy. Deep learning, especially using transformer models, has shown great potential in the recognition of gastrointestinal (GI) images according to self-attention. This study aims to establish an identification model of capsule endoscopy gastric structures to improve the clinical applicability of deep learning to endoscopic image recognition. METHODS: A total of 3343 wireless capsule endoscopy videos collected at Nanfang Hospital between 2011 and 2021 were used for unsupervised pretraining, while 2433 were for training and 118 were for validation. Fifteen upper GI structures were selected for quantifying the examination quality. We also conducted a comparison of the classification performance between the artificial intelligence model and endoscopists by the accuracy, sensitivity, specificity, and positive and negative predictive values. RESULTS: The transformer-based AI model reached a relatively high level of diagnostic accuracy in gastric structure recognition. Regarding the performance of identifying 15 upper GI structures, the AI model achieved a macroaverage accuracy of 99.6% (95% CI: 99.5-99.7), a macroaverage sensitivity of 96.4% (95% CI: 95.3-97.5), and a macroaverage specificity of 99.8% (95% CI: 99.7-99.9) and achieved a high level of interobserver agreement with endoscopists. CONCLUSIONS: The transformer-based AI model can accurately evaluate the gastric structure information of capsule endoscopy with the same performance as that of endoscopists, which will provide tremendous help for doctors in making a diagnosis from a large number of images and improve the efficiency of examination.

RUNX1 promotes angiogenesis in colorectal cancer by regulating the crosstalk between tumor cells and tumor associated macrophages.

Colorectal cancer (CRC) is a common malignancy worldwide. Angiogenesis and metastasis are the critical hallmarks of malignant tumor. Runt-related transcription factor 1 (RUNX1), an efficient transcription factor, facilitates CRC proliferation, metastasis and chemotherapy resistance. We aimed to investigate the RUNX1 mediated crosstalk between tumor cells and M2 polarized tumor associated macrophages (TAMs) in CRC, as well as its relationship with neoplastic angiogenesis. We found that RUNX1 recruited macrophages and induced M2 polarized TAMs in CRC by promoting the production of chemokine 2 (CCL2) and the activation of Hedgehog pathway. In addition, we found that the M2 macrophage-specific generated cytokine, platelet-derived growth factor (PDGF)-BB, promoted vessel formation both in vitro and vivo. PDGF-BB was also found to enhance the expression of RUNX1 in CRC cell lines, and promote its migration and invasion in vitro. A positive feedback loop of RUNX1 and PDGF-BB was thus formed. In conclusion, our data suggest that RUNX1 promotes CRC angiogenesis by regulating M2 macrophages during the complex crosstalk between tumor cells and TAMs. This observation provides a potential combined therapy strategy targeting RUNX1 and TAMs-related PDGF-BB in CRC.

DDX39B contributes to the proliferation of colorectal cancer through direct binding to CDK6/CCND1.

DDX39B (also called UAP56 or BAT1) which is a kind of DEAD-box family helicase plays pivotal roles in mRNA binding, splicing, and export. It has been found upregulated in many kinds of tumors as an oncogene. Nevertheless, the underlying molecular mechanisms of DDX39B in the proliferation of human colorectal cancer (CRC) remain fairly elusive. In our study, function experiments including the CCK8 and colony formation assay revealed that DDX39B facilitates CRC proliferation in vitro. DDX39B knockdown cells were administered for the orthotopic CRC tumor xenograft mouse model, after which tumor growth was monitored and immunohistochemistry (IHC) was performed to prove that DDX39B can also facilitates CRC proliferation in vivo. Flow cytometry demonstrated that DDX39B promotes the proliferation of CRC cells by driving the cell cycle from G0/G1 phase to the S phase. Mechanistically, RNA-binding protein immunoprecipitation-sequencing (RIP-seq) confirmed that DDX39B binds directly to the first exon of the CDK6/CCND1 pre-mRNA and upregulates their expression. Splicing experiments in vitro using a RT-PCR and gel electrophoresis assay confirmed that DDX39B promotes CDK6/CCND1 pre-mRNA splicing. Rescue experiments indicated that CDK6/CCND1 is a downstream effector of DDX39B-mediated CRC cell proliferation. Collectively, our results demonstrated that DDX39B and CDK6/CCND1 direct interactions serve as a CRC proliferation promoter, which can accelerate the G1/S phase transition to enhance CRC proliferation, and can offer novel and emerging treatment strategies targeting this cell proliferation-promoting gene.

Convolution neural network for the diagnosis of wireless capsule endoscopy: a systematic review and meta-analysis.

BACKGROUND: Wireless capsule endoscopy (WCE) is considered to be a powerful instrument for the diagnosis of intestine diseases. Convolution neural network (CNN) is a type of artificial intelligence that has the potential to assist the detection of WCE images. We aimed to perform a systematic review of the current research progress to the CNN application in WCE. METHODS: A search in PubMed, SinoMed, and Web of Science was conducted to collect all original publications about CNN implementation in WCE. Assessment of the risk of bias was performed by Quality Assessment of Diagnostic Accuracy Studies-2 risk list. Pooled sensitivity and specificity were calculated by an exact binominal rendition of the bivariate mixed-effects regression model. I2 was used for the evaluation of heterogeneity. RESULTS: 16 articles with 23 independent studies were included. CNN application to WCE was divided into detection on erosion/ulcer, gastrointestinal bleeding (GI bleeding), and polyps/cancer. The pooled sensitivity of CNN for erosion/ulcer is 0.96 [95% CI 0.91, 0.98], for GI bleeding is 0.97 (95% CI 0.93-0.99), and for polyps/cancer is 0.97 (95% CI 0.82-0.99). The corresponding specificity of CNN for erosion/ulcer is 0.97 (95% CI 0.93-0.99), for GI bleeding is 1.00 (95% CI 0.99-1.00), and for polyps/cancer is 0.98 (95% CI 0.92-0.99). CONCLUSION: Based on our meta-analysis, CNN-dependent diagnosis of erosion/ulcer, GI bleeding, and polyps/cancer approached a high-level performance because of its high sensitivity and specificity. Therefore, future perspective, CNN has the potential to become an important assistant for the diagnosis of WCE.

Identification of Immunological Biomarkers of Atopic Dermatitis by Integrated Analysis to Determine Molecular Targets for Diagnosis and Therapy.

PURPOSE: Atopic dermatitis (AD) is a common chronic inflammatory skin disorder associated with immune dysregulation and barrier dysfunction. In this study, we investigated immunological biomarkers for AD diagnosis and treatment using CIBERSORT to identify immune cell infiltration characteristics. PATIENTS AND METHODS: Common differentially expressed genes (DEGs) of lesioned (LS) vs non-lesioned (NL) groups were obtained from public datasets (GSE140684 and GSE99802). We performed functional enrichment analysis and selected hub genes from the protein-protein interaction (PPI) network. The hub genes were then subjected to transcription factor (TF), microRNA (miRNA), long non-coding RNA (lncRNA), drug interaction, and protein subcellular localization analyses. We also performed correlation analysis on differentially expressed immune cells, TFs, and hub genes. Receiver operating characteristic (ROC) curve analysis and binomial least absolute shrinkage and selection operator (LASSO) regression analysis were employed to assess the expression of hub genes in the GSE99802, GSE140684, GSE58558, GSE120721, and GSE36842 datasets. RESULTS: We identified 238 common DEGs and 25 hub genes. Additionally, we predicted TFs, miRNAs, lncRNA, drugs, and protein subcellular localizations. The proportions of activated dendritic cells (DCs) and CD4+ memory T cells were relatively high in the LS skin. Expression levels of the TF FOXC1 were negatively correlated with target genes and the abundance of two immune cell types. The LASSO model showed that GZMB, CXCL1, and CD274 are candidate diagnostic biomarkers. CONCLUSION: Our study suggests that downregulated expression of FOXC1 expression may enhance the levels of chemokines, chemokine receptors, T cell receptor signaling molecules, activating CD4+ memory T cells and DCs in AD.

Role of the Receptor for Advanced Glycation End Products in Heat Stress-Induced Endothelial Hyperpermeability in Acute Lung Injury.

OBJECTIVE: To study the role of the receptor for advanced glycation end products (RAGE) in endothelial barrier dysfunction induced by heat stress, to further explore the signal pathway by which RAGE contributes to heat-induced endothelia response, and thereby find a novel target for the clinical treatment of ALI (acute lung injury) induced by heatstroke. METHODS: This study established the animal model of heatstroke using RAGE knockout mice. We observed the role of RAGE in acute lung injury induced by heatstroke in mice by evaluating the leukocytes, neutrophils, and protein concentration in BALF (Bronchoalveolar lavage fluids), lung wet/dry ratio, histopathological changes, and the morphological ultrastructure of lung tissue and arterial blood gas analysis. To further study the mechanism, we established a heat stress model of HUVEC and concentrated on the role of RAGE and its signal pathway in the endothelial barrier dysfunction induced by heat stress, measuring Transendothelial electrical resistance (TEER) and western blot. RESULTS: RAGE played a key role in acute lung injury induced by heatstroke in mice. The mechanism C-Jun is located in the promoter region of the RAGE gene. C-Jun increased the RAGE protein expression while HSF1 suppressed RAGE protein expression. The overexpressed RAGE protein then increased HUVEC monolayer permeability by activating ERK and P38 MAPK under heat stress. CONCLUSION: This study indicates the critical role of RAGE in heat stress-induced endothelial hyperpermeability in acute lung injury and suggests that RAGE could be a potential therapeutic target in protecting patients against acute lung injury induced by heatstroke.