Clinical application of machine learning-based pathomics signature of gastric atrophy.

Background: The diagnosis of gastric atrophy is highly subjective, and we aimed to establish a model of gastric atrophy based on pathological features to improve diagnostic consistency. Methods: We retrospectively collected the HE-stained pathological slides of gastric biopsies and used CellProfiler software for image segmentation and feature extraction of ten representative images for each sample. Subsequently, we employed the Least absolute shrinkage and selection operator (LASSO) to select features and different machine learning (ML) algorithms to construct the diagnostic models for gastric atrophy. Results: We selected 289 gastric biopsy specimens for training, testing, and external validation. We extracted 464 pathological features and screened ten features by LASSO to establish the diagnostic model for moderate-to-severe atrophy. The range of area under the curve (AUC) for various machine learning algorithms was 0.835-1.000 in the training set, 0.786-0.949 in the testing set, and 0.689-0.818 in the external validation set. LR model had the highest AUC value, with 0.900 (95% CI: 0.852-0.947) in the training set, 0.901 (95% CI: 0.807-0.996) in the testing set, and 0.818 (95% CI: 0.714-0.923) in the external validation set. The atrophy pathological score based on the LR model was associated with endoscopic atrophy grading (Z=-2.478, P=0.013) and gastric cancer (GC) (OR=5.70, 95% CI: 2.63-12.33, P<0.001). Conclusion: The ML model based on pathological features could improve the diagnostic consistency of gastric atrophy, which is also associated with endoscopic atrophy grading and GC.

Quantitative Proteomics Analysis Reveals the Potential Role of lncRNA FTX in Endothelial Cells.

OBJECTIVE: This study investigated the role of long non-coding RNAs (lncRNAs) FTX in vascular endothelial cells (ECs). METHODS: Transfection of FTX/Sh-FTX with lentivirus was used to construct gain and loss of function cell models in human umbilical vein endothelial cells (HUVECs). Liquid chromatography-mass spectrometry was used for quantitative proteomics analysis of differentially expressed proteins (DEPs). Gene Ontology, Kyoto Encyclopedia of Genes and Genomes, and protein interaction analysis were further conducted to investigate the key molecules and pathways that respond to lncRNA-FTX. RESULTS: In the proteomics analysis, 3308 quantifiable proteins were identified, 64 proteins were upregulated and 103 were downregulated when lncRNA FTX was overexpressed. Additionally, 100 proteins were upregulated and 147 were downregulated when lncRNA FTX was knocked down. Functional clustering analysis of DEPs demonstrated that lncRNA FTX was involved in multiple biological processes. Among them, the expression of complement 3 (C3), cartilage oligomeric matrix protein (COMP), faciogenital dysplasia 6 (FGD6), and tissue inhibitor of metalloproteinase 1 (TIMP1) was significantly upregulated when lncRNA FTX was knocked down, and significantly downregulated when lncRNA FTX was overexpressed. They are associated with inflammation, collagen deposition, angiogenesis, and regulation of liver stem cell differentiation, which may be associated with the occurrence and development of liver fibrosis. CONCLUSIONS: The study demonstrated that lncRNA FTX might play a potential role in ECs and contribute to the development of liver fibrosis. Thus, FTX may be a promising target for the prevention or reversal of liver fibrosis.

Image-based AI diagnostic performance for fatty liver: a systematic review and meta-analysis.

BACKGROUND: The gold standard to diagnose fatty liver is pathology. Recently, image-based artificial intelligence (AI) has been found to have high diagnostic performance. We systematically reviewed studies of image-based AI in the diagnosis of fatty liver. METHODS: We searched the Cochrane Library, Pubmed, Embase and assessed the quality of included studies by QUADAS-AI. The pooled sensitivity, specificity, negative likelihood ratio (NLR), positive likelihood ratio (PLR), and diagnostic odds ratio (DOR) were calculated using a random effects model. Summary receiver operating characteristic curves (SROC) were generated to identify the diagnostic accuracy of AI models. RESULTS: 15 studies were selected in our meta-analysis. Pooled sensitivity and specificity were 92% (95% CI: 90-93%) and 94% (95% CI: 93-96%), PLR and NLR were 12.67 (95% CI: 7.65-20.98) and 0.09 (95% CI: 0.06-0.13), DOR was 182.36 (95% CI: 94.85-350.61). After subgroup analysis by AI algorithm (conventional machine learning/deep learning), region, reference (US, MRI or pathology), imaging techniques (MRI or US) and transfer learning, the model also demonstrated acceptable diagnostic efficacy. CONCLUSION: AI has satisfactory performance in the diagnosis of fatty liver by medical imaging. The integration of AI into imaging devices may produce effective diagnostic tools, but more high-quality studies are needed for further evaluation.

Therapeutic effects and potential mechanisms of endoscopic submucosal injection of mesenchymal stem cells on chronic atrophic gastritis.

Previous studies have demonstrated the rejuvenating and restorative actions of mesenchymal stem cells (MSCs) in multiple diseases, but their role in reversing chronic atrophic gastritis (CAG) is not well understood owing to their low efficiency in homing to the stomach. In this work, we investigated the therapeutic effect of umbilical cord-derived MSCs (UC-MSCs) on CAG by endoscopic submucosal injection and preliminarily explored possible mechanisms in vitro. MSCs and normal saline (NS) were injected into the submucosa of the stomach in randomly grouped CAG rabbits. Therapeutic effects on serum indices and histopathology of the gastric mucosa were analyzed in vivo at 30 and 60 days after MSCs injection. GES-1 cells were co-cultured with MSCs in vitro using a Transwell system and cell viability, proliferation, and migration ability were detected. Additionally, in view of the potential mechanisms, the relative protein expression levels of apoptosis, autophagy and inflammation in vitro were explored by Western Blotting. We found that submucosal injection of MSCs up-regulated serum indices (G-17, PGI and PGI/PGII) and alleviated histopathological damage to the gastric mucosa in CAG rabbits. Co-culture of GES-1 cells with MSCs improved cell viability, proliferation, and migration ability, while suppressing apoptosis. We also observed a reduction in the expression of apoptosis indicators, including Bax and cleaved caspase-3, in GES-1 cells after co-culture with MSCs in vitro. Our findings suggest that submucosal injection of MSCs is a promising approach for reversing CAG, and attenuating apoptosis plays a potential role in this process.

Biomimetic Boron Nitride Nanoparticles for Targeted Drug Delivery and Enhanced Antitumor Activity.

Boron nitride nanomaterials are being increasingly recognized as vehicles for cancer drug delivery that increase drug loading and control drug release because of their excellent physicochemical properties and biocompatibility. However, these nanoparticles are often cleared rapidly by the immune system and have poor tumor targeting effects. As a result, biomimetic nanotechnology has emerged to address these challenges in recent times. Cell-derived biomimetic carriers have the characteristics of good biocompatibility, long circulation time, and strong targeting ability. Here, we report a biomimetic nanoplatform (CM@BN/DOX) prepared by encapsulating boron nitride nanoparticles (BN) and doxorubicin (DOX) together using cancer cell membrane (CCM) for targeted drug delivery and tumor therapy. The CM@BN/DOX nanoparticles (NPs) were able to target cancer cells of the same type on its own initiative through homologous targeting of cancer cell membranes. This led to a remarkable increase in cellular uptake. In vitro simulation of an acidic tumor microenvironment could effectively promote drug release from CM@BN/DOX. Furthermore, the CM@BN/DOX complex exhibited an excellent inhibitory effect against homotypic cancer cells. These findings suggest that CM@BN/DOX are promising in targeted drug delivery and potentially personalized therapy against their homologous tumor.

The effect of trait anxiety on the time course of self-relevant processing: Evidence from the perceptual matching task.

Studies have widely reported that trait anxiety is associated with a range of cognitive biases toward external negative emotional stimuli. However, few studies have examined whether trait anxiety modulates intrinsic self-relevant processing. This study investigated the electrophysiological mechanism underlying trait anxiety's modulating effect on self-relevant processing. Event-related potentials (ERPs) were recorded while participants performed a perceptual matching task that assigned an arbitrary geometric shape to an association with a "self" or "non-self" label. Results showed larger N1 amplitudes under self-association than under friend-association conditions, and smaller P2 amplitudes for self- than for stranger-association conditions in individuals with high trait anxiety. However, these self-biases in the N1 and P2 stages were not observed in those with low trait anxiety until the later N2 stage, in which the self-association condition provoked smaller N2 amplitudes than the stranger-association condition. In addition, both high and low trait anxiety individuals showed larger P3 amplitudes for the self-association condition than for the friend- and stranger-association conditions. These findings suggest that, although both high and low trait anxiety individuals showed self-bias, high trait anxiety individuals distinguished between self-relevant and non-self-relevant stimuli at an earlier stage, which may reflect hypervigilance to self-relevant stimuli.