BACKGROUND:The local immune microenvironment plays an important regulatory role in the process of bone formation,and the immune system is intricately linked to the skeletal system.OBJECTIVE:To systematically review the promotion of bone regeneration from three aspects:immune cell regulation of microenvironment,regulation of immune response by small extracellular vesicles,and induction of immune response by bone biomaterials,and to elucidate the immune regulatory mechanisms involved in bone regeneration.METHODS:Relevant literature was retrieved from PubMed,CNKI,WanFang Database,and VIP Database,using the search terms of"osteoimmunology,immune microenvironment,small extracellular vesicles,bone regeneration,bone tissue repair,biomaterials,and tissue engineering"in English and Chinese.Repeat and irrelevant literature was screened and removed,and 92 articles that met the criteria were selected for intensive reading and review.RESULTS AND CONCLUSION:Multiple immune cells and bone cells are in the same microenvironment,and immune cells can regulate the differentiation and activity of bone cells,collectively forming an immune microenvironment that affects bone regeneration.Neutrophils can significantly reduce local inflammatory responses in the early stages of bone injury,creating a favorable microenvironment for bone regeneration.M1 macrophages can clear foreign bodies and reduce early inflammatory responses,while M2 macrophages can promote the expression of osteogenic markers and factors,playing an important role in the repair process of bone injury.B cells and T cells can directly or indirectly affect the generation and activity of osteoblasts and osteoclasts,regulate bone metabolism,and promote bone regeneration.Extracellular vesicles of small cells regulate the local immune microenvironment through paracrine secretion,promoting bone formation and angiogenesis at the site of bone injury.The metal ions,surface hydrophilicity,porosity,pore size,surface morphology,and surface roughness on the surface of biomaterials can directly regulate local immune responses,and have anti-inflammatory,angiogenic,and osteogenic effects,thereby accelerating bone regeneration.

BACKGROUND:The local immune microenvironment plays an important regulatory role in the process of bone formation,and the immune system is intricately linked to the skeletal system.OBJECTIVE:To systematically review the promotion of bone regeneration from three aspects:immune cell regulation of microenvironment,regulation of immune response by small extracellular vesicles,and induction of immune response by bone biomaterials,and to elucidate the immune regulatory mechanisms involved in bone regeneration.METHODS:Relevant literature was retrieved from PubMed,CNKI,WanFang Database,and VIP Database,using the search terms of"osteoimmunology,immune microenvironment,small extracellular vesicles,bone regeneration,bone tissue repair,biomaterials,and tissue engineering"in English and Chinese.Repeat and irrelevant literature was screened and removed,and 92 articles that met the criteria were selected for intensive reading and review.RESULTS AND CONCLUSION:Multiple immune cells and bone cells are in the same microenvironment,and immune cells can regulate the differentiation and activity of bone cells,collectively forming an immune microenvironment that affects bone regeneration.Neutrophils can significantly reduce local inflammatory responses in the early stages of bone injury,creating a favorable microenvironment for bone regeneration.M1 macrophages can clear foreign bodies and reduce early inflammatory responses,while M2 macrophages can promote the expression of osteogenic markers and factors,playing an important role in the repair process of bone injury.B cells and T cells can directly or indirectly affect the generation and activity of osteoblasts and osteoclasts,regulate bone metabolism,and promote bone regeneration.Extracellular vesicles of small cells regulate the local immune microenvironment through paracrine secretion,promoting bone formation and angiogenesis at the site of bone injury.The metal ions,surface hydrophilicity,porosity,pore size,surface morphology,and surface roughness on the surface of biomaterials can directly regulate local immune responses,and have anti-inflammatory,angiogenic,and osteogenic effects,thereby accelerating bone regeneration.

AIM: To investigate the potential inhibitory effect of pterostilbene on the endothelial-to-mesenchymal transition(EndMT)induced by high glucose conditions in human retinal microvascular endothelial cells(HRMECs).METHODS: The optimal concentration of pterostilbene for treating HRMECs was determined using the CCK-8 assay, with 12.5 and 25 μmol/L concentrations selected for subsequent experiments. Four experimental groups were established: control group, high glucose group, high glucose combined with 12.5 μmol/L pterostilbene treatment group, and high glucose combined with 25 μmol/L pterostilbene treatment group. The expression levels of HDAC7 and EndMT-associated markers were detected via Western blot analysis. Cell migration ability was assessed using Transwell migration assays and scratch wound healing tests, while vasculogenic capability was evaluated through tube formation assays.RESULTS: The CCK-8 assay revealed that pterostilbene at a concentration of 22.07 μmol/L inhibited 50% of cell viability in HRMECs. Western blot analysis demonstrated that compared with the control group, the expression levels of HDAC7, ZEB1, Vimentin, and Snail were significantly upregulated in HRMECs cultured in high glucose(all P<0.01), while the expressions of VE-cadherin and CD31 were significantly reduced(all P<0.01). Compared to the high glucose group, the treatment with 12.5 and 25 μmol/L pterostilbene significantly reduced the expression of HDAC7, ZEB1, Vimentin, and Snail under high glucose conditions(all P<0.01). Notably, 25 μmol/L pterostilbene enhanced the expression of VE-cadherin and CD31(all P<0.01). Scratch wound healing tests revealed that HRMECs treated with high glucose exhibited a significantly increased cell migration rate compared to the control group(P<0.05), while the application of 25 μmol/L pterostilbene significantly suppressed HRMECs migration under high glucose conditions(P<0.01). Transwell migration assays demonstrated that the cell migration rate in the high glucose group was significantly higher than that in the control group(P<0.01), with cell migration rate markedly reduced following treatment with both of 12.5 and 25 μmol/L pterostilbene(all P<0.01). The tube formation assay revealed that the ability of HRMECs to form tubular structures was significantly enhanced under high glucose conditions(P<0.01), and both 12.5 and 25 μmol/L of pterostilbene effectively inhibited this effect(all P<0.01).CONCLUSION: Pterostilbene can inhibit HDAC7 expression, suppress EndMT-mediated migration of HRMECs, and impair tube formation under high-glucose conditions.

AIM: To investigate the potential inhibitory effect of pterostilbene on the endothelial-to-mesenchymal transition(EndMT)induced by high glucose conditions in human retinal microvascular endothelial cells(HRMECs).METHODS: The optimal concentration of pterostilbene for treating HRMECs was determined using the CCK-8 assay, with 12.5 and 25 μmol/L concentrations selected for subsequent experiments. Four experimental groups were established: control group, high glucose group, high glucose combined with 12.5 μmol/L pterostilbene treatment group, and high glucose combined with 25 μmol/L pterostilbene treatment group. The expression levels of HDAC7 and EndMT-associated markers were detected via Western blot analysis. Cell migration ability was assessed using Transwell migration assays and scratch wound healing tests, while vasculogenic capability was evaluated through tube formation assays.RESULTS: The CCK-8 assay revealed that pterostilbene at a concentration of 22.07 μmol/L inhibited 50% of cell viability in HRMECs. Western blot analysis demonstrated that compared with the control group, the expression levels of HDAC7, ZEB1, Vimentin, and Snail were significantly upregulated in HRMECs cultured in high glucose(all P<0.01), while the expressions of VE-cadherin and CD31 were significantly reduced(all P<0.01). Compared to the high glucose group, the treatment with 12.5 and 25 μmol/L pterostilbene significantly reduced the expression of HDAC7, ZEB1, Vimentin, and Snail under high glucose conditions(all P<0.01). Notably, 25 μmol/L pterostilbene enhanced the expression of VE-cadherin and CD31(all P<0.01). Scratch wound healing tests revealed that HRMECs treated with high glucose exhibited a significantly increased cell migration rate compared to the control group(P<0.05), while the application of 25 μmol/L pterostilbene significantly suppressed HRMECs migration under high glucose conditions(P<0.01). Transwell migration assays demonstrated that the cell migration rate in the high glucose group was significantly higher than that in the control group(P<0.01), with cell migration rate markedly reduced following treatment with both of 12.5 and 25 μmol/L pterostilbene(all P<0.01). The tube formation assay revealed that the ability of HRMECs to form tubular structures was significantly enhanced under high glucose conditions(P<0.01), and both 12.5 and 25 μmol/L of pterostilbene effectively inhibited this effect(all P<0.01).CONCLUSION: Pterostilbene can inhibit HDAC7 expression, suppress EndMT-mediated migration of HRMECs, and impair tube formation under high-glucose conditions.

Osteoporosis is a common bone metabolic disease， which is mainly characterized by the decrease in the number of bone trabeculae and the destruction of bone tissue microstructure， leading to increased bone fragility and fracture risks. This disease is common in postmenopausal women， elderly men， diabetes patients， and obese people. Due to the lack of awareness to prevent bone losses and the limitations of bone mass measurement methods， osteoporosis is only concerned when there are serious complications， which imposes a heavy burden on both patients and medical resources. Oxidative stress refers to the excessive production of highly active molecules such as reactive oxygen species and reactive nitrogen in the body subjected to harmful stimuli， leading to the imbalance between the oxidative and antioxidant systems and causing oxidative damage. Studies have shown that oxidative stress can increase the generation and activity of osteoclasts and inhibit the differentiation of osteoblasts， thus playing a role in the occurrence and development of osteoporosis. Traditional Chinese medicine （TCM） is considered an effective antioxidant that can alleviate oxidative stress-induced osteoporosis by regulating a variety of signaling pathways. Studies have shown that TCM can alleviate oxidative stress and promote bone angiogenesis and osteogenesis by regulating the phosphatidylinositol 3-kinase （PI3K）/protein kinase B （Akt）， nuclear factor-kappa B， and nuclear factor erythroid 2-related factor （Nrf2） signaling pathways. TCM alleviates oxidative stress and promotes osteogenesis by regulating the Nrf2， PI3K/Akt/mammalian target of rapamycin， and secreted glycoprotein Wnt/β-catenin signaling pathways. In addition， TCM regulates NF-κB， mitogen-activated protein kinase， and receptor activator of nuclear factor kappa B （RANK）/RANK ligand/osteoprotegerin signaling pathway to alleviate excessive bone resorption induced by oxidative stress. This paper systematically summarizes the literature on the prevention and treatment of osteoporosis by TCM or its active ingredients via the above-mentioned signaling pathways to reduce oxidative stress in recent years. It briefs the possible molecular mechanisms of oxidative stress regulation-related signaling pathways to cause osteoporosis. In addition， this paper discusses the effects and mechanisms of TCM on bone angiogenesis， osteogenesis， and bone resorption by reducing oxidative stress through the regulation of related signaling pathways， aiming to provide a theoretical basis for the research and clinical treatment of osteoporosis.

Objective To investigate the suitable habitats of Phlebotomus chinensis in Gansu Province, so as provide insights into effective management of mountain-type zoonotic visceral leishmaniasis (MT-ZVL). Methods The geographical coordinates of locations where MT-ZVL cases were reported were retrieved in Gansu Province from 2015 to 2023, and data pertaining to 26 environmental variables were captured, including 19 climatic variables (annual mean temperature, mean diurnal range, isothermality, temperature seasonality, maximum temperature of the warmest month, minimum temperature of the coldest month, temperature annual range, mean temperature of the wettest quarter, mean temperature of the driest quarter, mean temperature of the warmest quarter, mean temperature of the coldest quarter, annual precipitation, precipitation of the wettest month, precipitation of the driest month, precipitation seasonality, precipitation of the wettest quarter, precipitation of the driest quarter, precipitation of the warmest quarter, and precipitation of the coldest quarter), five geographical variables (elevation, annual normalized difference vegetation index, vegetation type, landform type and land use type), and two population and economic variables (population distribution and gross domestic product). Twelve species distribution models were built using the biomod2 package in R project, including surface range envelope (SRE) model, generalized linear model (GLM), generalized additive model (GAM), multivariate adaptive regression splines (MARS) model, generalized boosted model (GBM), classification tree analysis (CTA) model, flexible discriminant analysis (FDA) model, maximum entropy (MaxEnt) model, optimized maximum entropy (MAXNET) model, artificial neural network (ANN) model, random forest (RF) model, and extreme gradient boosting (XGBOOST) model. The performance of 12 models was evaluated using the area under the receiver operating characteristic curve (AUC), true skill statistics (TSS), and Kappa coefficient, and single models with high performance was selected to build the optimal ensemble models. Factors affecting the survival of Ph. chinensis were identified based on climatic, geographical, population and economic variables. In addition, the suitable distribution areas of Ph. chinensis were predicted in Gansu Province under shared socioeconomic pathway 126 (SSP126), SSP370 and SSP585 scenarios based on climatic data during the period from 1991 to 2020, from 2041 to 2060 (2050s), and from 2081 to 2100 (2090s) . Results A total of 11 species distribution models were successfully built for prediction of potential distribution areas of Ph. chinensis in Gansu Province, and the RF model had the highest predictive accuracy (AUC = 0.998). The ensemble model built based on the RF model, XGBOOST model, GLM, and MARS model had an increased predictive accuracy (AUC = 0.999) relative to single models. Among the 26 environmental factors, precipitation of the wettest quarter (12.00%), maximum temperature of the warmest month (11.58%), and annual normalized difference vegetation index (11.29%) had the greatest contributions to suitable habitats distribution of Ph. sinensis. Under the climatic conditions from 1991 to 2020, the potential suitable habitat area for Ph. chinensis in Gansu Province was approximately 5.80 × 10⁴ km², of which the highly suitable area was 1.42 × 10⁴ km², and primarily concentrated in the southernmost region of Gansu Province. By the 2050s, the unsuitable and lowly suitable areas for Ph. chinensis in Gansu Province had decreased by varying degrees compared to that of 1991 to 2020 period, while the moderately and highly suitable areas exhibited expansion and migration. By the 2090s, under the SSP126 scenario, the suitable habitat area for Ph. chinensis increased significantly, and under the SSP585 scenario, the highly suitable areas transformed into extremely suitable areas, also showing substantial growth. Future global warming is conducive to the survival and reproduction of Ph. chinensis. From the 2050s to the 2090s, the highly suitable areas for Ph. chinensis in Gansu Province will be projected to expand northward. Under the SSP126 scenario, the suitable habitat area for Ph. chinensis in Gansu Province is expected to increase by 194.75% and 204.79% in the 2050s and 2090s, respectively, compared to that of the 1991 to 2020 period. Under the SSP370 scenario, the moderately and highly suitable areas will be projected to increase by 164.40% and 209.03% in the 2050s and 2090s, respectively, while under the SSP585 scenario, they are expected to increase by 195.98% and 211.66%, respectively. Conclusions The distribution of potential suitable habitats of Ph. sinensis gradually shifts with climatic changes. Intensified surveillance and management of Ph. sinensis is recommended in central and eastern parts of Gansu Province to support early warning of MT-ZVL.

Objective:To develop a deep learning-based system for real-time recognition and classification of portal hypertensive gastropathy (PHG) and evaluate its ability to assist junior endoscopists.Methods:A total of 2 848 gastroscopy images from 832 patients with liver cirrhosis were selected from Digestive Endoscopy Center databases of Renmin Hospital of Wuhan University, Wuhan Hospital of Traditional Chinese and Western Medicine, and the Second Hospital of Jingzhou from January 2015 to October 2023. This system referred to 3 endoscopic features of Baveno Ⅱ scoring system. Three models were developed respectively for gastric antral vascular ectasia (GAVE), mosaic-like pattern (MLP), and red marks (RM). The specific classification references were as follows: (1) GAVE model: 0 no, 1 yes; (2) MLP model: 0 no, 1 mild, 2 severe; (3) RM model: 0 no, 1 isolated, 2 fused. The classification results for endoscopic characteristics of PHG of 3 endoscopy experts were taken as the gold standard. The yolov8-m model was used for training. The training dataset, validation dataset, and test dataset were allocated at a ratio of 8∶1∶1. The test dataset was used to evaluate the performance of models and their auxiliary effects on endoscopists. The accuracy, recall, precision, specificity and Kappa coefficient were calculated. Results:The accuracy, recall, specificity of GAVE model were 96.0% (48/50), 87.5% (7/8) and 97.6% (41/42). There was no significant difference between its accuracy and the gold standard ( χ2=316.226, P=1.000). The precision of GAVE1 and GAVE0 were 87.5% (7/8) and 97.6% (41/42) respectively. The accuracy of MLP model was 84.1% (132/157), and there was no significant difference compared with the gold standard ( χ2=3.286, P=0.193). The precision and recall of MLP2 were 88.2% (15/17) and 75.0% (15/20). The precision and recall of MLP1 were 77.9% (60/77) and 88.2% (60/68). The precision and recall of MLP0 were 90.5% (57/63) and 82.6% (57/69). The accuracy of RM model was 87.9% (123/140), and there was no significant difference compared with the gold standard ( χ2=2.891, P=0.409). The precision and recall of RM2 were 94.7% (18/19) and 78.3% (18/23). The precision and recall of RM1 were 72.2% (26/36) and 81.3% (26/32). The precision and recall of RM0 were 92.9% (79/85) and 92.9% (79/85). The mean accuracy of the three junior endoscopists, with and without the assistance of the GAVE model, MLP model, and RM model, respectively increased from 95.3% to 99.3%, from 83.9% to 91.9%, and from 81.9% to 83.1%. The overall consistency analysis of the 3 junior endoscopists with the gold standard indicated that the consistency of the GAVE model before and after assistance was extremely strong (both an overall Kappa of 1.000); the consistency before assistance of the MLP model was moderate (with an overall Kappa of 0.601), which increased to extremely strong after assistance (with an overall Kappa of 0.964); and the consistency of the RM model before and after assistance was also relatively strong (with an overall Kappa of 0.792 before and 0.798 after). Conclusion:The deep learning system accurately identifies and classifies PHG features and significantly enhances diagnostic performance of junior endoscopists.

BACKGROUND:Gegenmaqi prescription has a good effect on periarthritis of shoulder combined with type 2 diabetes and has a good application prospect,but the specific mechanism is not clear.OBJECTIVE:To explore the action mechanism of Gegenmaqi prescription on periarthritis of shoulder and type 2 diabetes by network pharmacology,molecular docking,and molecular dynamics.METHODS:The active components and protein targets of Gegenmaqi prescription were retrieved from the Traditional Chinese Medicine System Pharmacology database and analysis platform,referred to as TCMSP jointly established by the Shanghai Institute of Materia Medica,Chinese Academy of Sciences and the Institute of Chinese Materia Medica,and China Academy of Chinese Medical Sciences in 2013.Genecards created by Professor Doron Lancet's team at the Weizmann Institute of Science in Israel in 1997,Drugbank created by scientists at the University of Alberta in Canada in 2006,and the OMIM database established by Dr.Victor A.McKusick's team at Johns Hopkins University in the United States in 1966 were used to search the disease protein targets of periarthritis of shoulder and type 2 diabetes,and the intersection targets were obtained based on the WeChat online tool.The protein-protein interaction network was constructed based on the STRING database created in 2000 by Peer Bork's team at the European Bioinformatics Institute(EMBL),and the protein-protein interaction relationship was analyzed.The core targets were screened according to the degree value.The intersection targets were subjected to GO and KEGG enrichment analyses.Finally,molecular docking and molecular dynamics simulation were used to verify the binding of key components to key targets.RESULTS AND CONCLUSION:(1)One hundred and forty-two active ingredients of Gegenmaqi prescription were obtained,including 65 intersections between component targets and disease targets,5 key active ingredients(β-sitosterol,stigmasterol,kaempferol,quercetin,and formononetin),and 5 key targets(AKT1,tumor necrosis factor,interleukin-10,JUN,and TP53).(2)GO function enrichment included 508 items,390 biological process items,77 molecular function items and 41 cell component items.KEGG pathway analysis showed 146 pathways,mainly involving advanced glycation end products receptor signaling pathway,lipid and atherosclerosis signaling pathway,tumor necrosis factor signaling pathway,and interleukin-17 signaling pathway.(3)Molecular docking showed that the key components and key targets had good binding activity.Molecular dynamics simulation showed that β-sitosterol had stable interactions with AKT1,tumor necrosis factor and interleukin 10.(4)Gegenmaqi prescription has been comprehensively studied,and the material basis of its pharmacological effect has been primarily clarified.It is predicted that Gegenmaqi prescription can treat periarthritis of shoulder combined with type 2 diabetes through multi-components,multi-targets,and multi-pathways to exert anti-inflammatory and regulate insulin secretion.

BACKGROUND:Gegenmaqi prescription has a good effect on periarthritis of shoulder combined with type 2 diabetes and has a good application prospect,but the specific mechanism is not clear.OBJECTIVE:To explore the action mechanism of Gegenmaqi prescription on periarthritis of shoulder and type 2 diabetes by network pharmacology,molecular docking,and molecular dynamics.METHODS:The active components and protein targets of Gegenmaqi prescription were retrieved from the Traditional Chinese Medicine System Pharmacology database and analysis platform,referred to as TCMSP jointly established by the Shanghai Institute of Materia Medica,Chinese Academy of Sciences and the Institute of Chinese Materia Medica,and China Academy of Chinese Medical Sciences in 2013.Genecards created by Professor Doron Lancet's team at the Weizmann Institute of Science in Israel in 1997,Drugbank created by scientists at the University of Alberta in Canada in 2006,and the OMIM database established by Dr.Victor A.McKusick's team at Johns Hopkins University in the United States in 1966 were used to search the disease protein targets of periarthritis of shoulder and type 2 diabetes,and the intersection targets were obtained based on the WeChat online tool.The protein-protein interaction network was constructed based on the STRING database created in 2000 by Peer Bork's team at the European Bioinformatics Institute(EMBL),and the protein-protein interaction relationship was analyzed.The core targets were screened according to the degree value.The intersection targets were subjected to GO and KEGG enrichment analyses.Finally,molecular docking and molecular dynamics simulation were used to verify the binding of key components to key targets.RESULTS AND CONCLUSION:(1)One hundred and forty-two active ingredients of Gegenmaqi prescription were obtained,including 65 intersections between component targets and disease targets,5 key active ingredients(β-sitosterol,stigmasterol,kaempferol,quercetin,and formononetin),and 5 key targets(AKT1,tumor necrosis factor,interleukin-10,JUN,and TP53).(2)GO function enrichment included 508 items,390 biological process items,77 molecular function items and 41 cell component items.KEGG pathway analysis showed 146 pathways,mainly involving advanced glycation end products receptor signaling pathway,lipid and atherosclerosis signaling pathway,tumor necrosis factor signaling pathway,and interleukin-17 signaling pathway.(3)Molecular docking showed that the key components and key targets had good binding activity.Molecular dynamics simulation showed that β-sitosterol had stable interactions with AKT1,tumor necrosis factor and interleukin 10.(4)Gegenmaqi prescription has been comprehensively studied,and the material basis of its pharmacological effect has been primarily clarified.It is predicted that Gegenmaqi prescription can treat periarthritis of shoulder combined with type 2 diabetes through multi-components,multi-targets,and multi-pathways to exert anti-inflammatory and regulate insulin secretion.

Objective:To develop a deep learning-based system for real-time recognition and classification of portal hypertensive gastropathy (PHG) and evaluate its ability to assist junior endoscopists.Methods:A total of 2 848 gastroscopy images from 832 patients with liver cirrhosis were selected from Digestive Endoscopy Center databases of Renmin Hospital of Wuhan University, Wuhan Hospital of Traditional Chinese and Western Medicine, and the Second Hospital of Jingzhou from January 2015 to October 2023. This system referred to 3 endoscopic features of Baveno Ⅱ scoring system. Three models were developed respectively for gastric antral vascular ectasia (GAVE), mosaic-like pattern (MLP), and red marks (RM). The specific classification references were as follows: (1) GAVE model: 0 no, 1 yes; (2) MLP model: 0 no, 1 mild, 2 severe; (3) RM model: 0 no, 1 isolated, 2 fused. The classification results for endoscopic characteristics of PHG of 3 endoscopy experts were taken as the gold standard. The yolov8-m model was used for training. The training dataset, validation dataset, and test dataset were allocated at a ratio of 8∶1∶1. The test dataset was used to evaluate the performance of models and their auxiliary effects on endoscopists. The accuracy, recall, precision, specificity and Kappa coefficient were calculated. Results:The accuracy, recall, specificity of GAVE model were 96.0% (48/50), 87.5% (7/8) and 97.6% (41/42). There was no significant difference between its accuracy and the gold standard ( χ2=316.226, P=1.000). The precision of GAVE1 and GAVE0 were 87.5% (7/8) and 97.6% (41/42) respectively. The accuracy of MLP model was 84.1% (132/157), and there was no significant difference compared with the gold standard ( χ2=3.286, P=0.193). The precision and recall of MLP2 were 88.2% (15/17) and 75.0% (15/20). The precision and recall of MLP1 were 77.9% (60/77) and 88.2% (60/68). The precision and recall of MLP0 were 90.5% (57/63) and 82.6% (57/69). The accuracy of RM model was 87.9% (123/140), and there was no significant difference compared with the gold standard ( χ2=2.891, P=0.409). The precision and recall of RM2 were 94.7% (18/19) and 78.3% (18/23). The precision and recall of RM1 were 72.2% (26/36) and 81.3% (26/32). The precision and recall of RM0 were 92.9% (79/85) and 92.9% (79/85). The mean accuracy of the three junior endoscopists, with and without the assistance of the GAVE model, MLP model, and RM model, respectively increased from 95.3% to 99.3%, from 83.9% to 91.9%, and from 81.9% to 83.1%. The overall consistency analysis of the 3 junior endoscopists with the gold standard indicated that the consistency of the GAVE model before and after assistance was extremely strong (both an overall Kappa of 1.000); the consistency before assistance of the MLP model was moderate (with an overall Kappa of 0.601), which increased to extremely strong after assistance (with an overall Kappa of 0.964); and the consistency of the RM model before and after assistance was also relatively strong (with an overall Kappa of 0.792 before and 0.798 after). Conclusion:The deep learning system accurately identifies and classifies PHG features and significantly enhances diagnostic performance of junior endoscopists.