Objective:To analyze measurement results of serum allergen-specific immunoglobulin E (IgE) in patients with eczema/dermatitis.Methods:A retrospective analysis was conducted on serum allergen-specific immunoglobulin E (IgE) levels in 3 051 patients with eczema/dermatitis, who visited the allergy clinic of Huashan Hospital from April 1, 2021 to March 31, 2022. The serum allergen-specific IgE level was detected by using the Phadia allergen detection system, and positive rates of allergens were calculated to determine common inhaled allergens and food allergens in patients with eczema/dermatitis. Comparisons of enumeration data between groups were performed by chi-square test.Results:Among the 3 051 patients with eczema/dermatitis, there were 1 412 with atopic dermatitis and 1 639 were other eczema/dermatitis. Detection of serum allergen-specific IgE showed that 1 629 (53%) patients were positive for allergens, and the number of positive allergen-specific IgEs in each patient was 3.0 ± 1.6. The top 3 common inhaled allergens in patients with eczema/dermatitis were Dermatophagoides farinae (904/1 522, 59%) , Dermatophagoides pteronyssinus (891/1 513, 59%) and Alternaria alternata (206/1 068, 19%) , and the top 3 common food allergens were shrimps (251/1 432, 18%) , egg white (165/992, 17%) and cow milk (149/994, 15%) . Among the 3 051 patients, 25 (1%) were aged < 2 years, 571 (19%) aged 2 - 12 years, 285 (9%) aged 12 - 18 years, and 2 170 (71%) were aged > 18 years. The most common food allergens were both egg white in the age groups of < 2 years and 2 -12 years (77%, 37%, respectively) , and were both shrimps in the age groups of 12 - 18 years and > 18 years (31%, 17%, respectively) . Dermatophagoides pteronyssinus and Dermatophagoides farina were the top 2 common inhaled allergens in all age groups, with the positive rate ranging from 36% to 84%; in addition, the positive rate of molds was relatively high in the age group of 2 - 12 years (mold mixture: 37%; Alternaria alternata: 27%) . From April 2021 to March 2022, the positive rate of outdoor allergens ranged from 10% to 15% among outpatients in every month; the positive rates of tree pollen and grass pollen increased from April 2021, and peaked in October 2021. The patients with atopic dermatitis showed a significantly increased positive rate of allergens (73%) compared with those with other eczema/dermatitis (37%, χ2 = 389.36, P<0.001) , and the rank of common allergens in the patients with atopic dermatitis was basically the same as that in those with eczema/dermatitis. Conclusions:The common allergens were Dermatophagoides farina, Dermatophagoides pteronyssinus and Alternaria alternata in the patients with eczema/dermatitis. Food allergy was more common in infant patients, and inhalation allergy was more common in child, adolescent and adult patients. The positive rate of allergen-specific IgEs was markedly higher in the patients with atopic dermatitis than in those with other eczema/dermatitis.

ObjectiveTo investigate the effects of Huashi Runzao prescription （HRP） on the histopathological injury and function of submandibular gland in naive non-obese diabetic （NOD/Ltj） mouse model of Sjögren's syndrome （SS） and its regulatory effect on aquaporin 5 （AQP5） expression in submandibular gland cells. MethodThe SS model was induced in NOD/Ltj mice. The NOD/Ltj female mice aged nine weeks were selected and randomly assigned into model group，HRP group （7.15 g·kg^-1·d^-1），and hydroxychloroquine （HCQ） group （1.30 g·kg^-1·d^-1）， and female BALB/c mice in the same age were selected and assigned into the normal group， with six mice in each group. Drug intervention lasted eight weeks. The water consumption and salivary flow rate （SFR） of each group were recorded. The pathological staining results of the submandibular gland of mice in each group were observed and scored. AQP5 expression was determined by immunohistochemistry （IHC） and Western blot. ResultCompared with the normal group， the model group showed increased water consumption （P<0.05） and reduced SFR （P<0.05）. Compared with the model group， the HRP group showed decreased water consumption （P<0.05） and increased SFR （P<0.05）， and the HCQ group showed increased SFR （P<0.05）. In terms of histopathological results of the submandibular gland，compared with the normal group，the model group showed increased pathological score， number of lymphocyte infiltration foci，and percentage of lymphatic infiltration area （P<0.05）. Compared with the model group， the HRP group showed reduced pathological scores and number of lymphocyte infiltration foci （P<0.05）， and the HRP group and the HCQ group showed reduced percentage of lymphatic infiltration area（P<0.05）. The results of IHC and Western blot showed that compared with the normal group，the model group showed down-regulated expression level of AQP5 protein （P<0.05）， and compared with the model group and the HCQ group，the HRP group showed up-regulated expression level of AQP5 protein （P<0.05）. ConclusionHRP can improve the secretion function of submandibular gland acinous cells and glandular structure injury in SS model mice， and its mechanism may be related to the up-regulation of AQP5 protein expression level in submandibular gland cells.

ObjectiveTo observe the efficacy and safety of Huashi Runzao prescription for patients with primary Sjögren's syndrome （pSS） of combined dryness and dampness pattern. MethodA total of 105 eligible patients were randomized into the experimental group （65 cases） and control group （40 cases）， and they were respectively treated with Huashi Runzao prescription and hydroxychloroquine for 12 weeks. Visual Analogue Scale （VAS） was employed to assess the symptoms. The symptoms of dryness， fatigue， and pain， European League Against Rheumatism （EULAR） Sjögren's Syndrome Patient Reported Index （ESSPRI）， EULAR Sjögren's syndrome disease activity index （ESSDAI）， and immune inflammatory indicators before and after treatment were compared between the two groups， and adverse reactions were observed. ResultAfter treatment， the ESSPRI score was lower than that before treatment in the experimental groups （P<0.01） and was lower in the experimental group than in the control group （P<0.05）. The VAS scores of dry mouth， dry eyes， overall dryness， fatigue， and pain in the experimental group decreased compared with those before treatment （P<0.01）， and the experimental group had lower VAS scores of dry mouth and overall dryness than the control group （P<0.01）. After treatment， the ESSDAI score of both groups decreased compared with that before treatment （P<0.05， P<0.01）， but there was no significant difference between the groups. After treatment， the level of immunoglobulin M （IgM） decreased （P<0.01） and the level of complement C3 increased （P<0.01） in the experimental group， while the level of complement C3 decreased in the control group （P<0.05）. There was no significant difference in the laboratory indexes between groups. During the treatment， stomachache occurred to one case in the experimental group， which was alleviated after the treatment， and no adverse reaction was observed in the control group. According to the chi-square test， the occurrence of adverse reactions was insignificantly different between the two groups. ConclusionHuashi Runzao prescription can alleviate the symptoms of dryness， fatigue， and pain， and reduce disease activity without associated side effects in pSS patients with combined dampness and dryness pattern.

ObjectiveTo explore the characteristics of tongue manifestations and traditional Chinese medicine （TCM） syndromes in patients with primary Sjogren's syndrome （pSS） of different time. MethodpSS patients who visited TCM Department of Rheumatism in China-Japan Friendship Hospital from December 2018 to October 2021 were included and grouped according to the disease courses （short<5 years， medium 5-10 years， long > 10 years）. Chi-square test was used for comparison between three groups and Bonferroni method for multiple comparisons. In the pairwise comparison， P<0.017 indicated significant difference. Chi-square test was performed on the syndrome and specific tongue manifestations with inter-group differences to analyze the trend of tongue manifestations and TCM syndromes over time （interval： two years）. ResultA total of 193 pSS patients were enrolled， with 85 （44.0%） of short disease course， 69 （35.8%） of medium disease course， and 39 （20.2%） of long disease course. The common tongue manifestations were crimson tongue， fissured tongue， thin tongue， lack of fluid， and dry coating， which showed no significant difference among the three groups. Higher proportion of patients with light red tongue was observed in the group with short disease course than in group with medium disease course （χ²=6.407， P<0.017）. Higher proportions of patients with thick coating （χ²=6.784， P<0.017） and phlegm-dampness syndrome （χ²=11.545， P<0.017） and lower proportion of patients with Qi deficiency syndrome （χ²=12.706， P<0.017） were found in the group with short disease course than in the group with long disease course. Patients with medium （χ²=6.358， P<0.017） and long （χ²=8.279， P<0.017） disease course tended to have exfoliated coating compared with those with short disease course， and the proportion of patients with exfoliated coating rose and the proportion of patients with thick greasy coating decreased over time （P_trend<0.05）. In addition， the proportion of patients with phlegm-dampness syndrome decreased and that with Qi deficiency syndrome increased over time （P_trend<0.05）. ConclusionIn the early stage， patients with pSS often show both dryness and dampness， as manifested by the thick greasy coating and phlegm-dampness syndrome. In the medium and late stage， patients often have Qi-Yin deficiency， as evidenced by exfoliated coating and Qi deficiency. In the clinical practice， medicines should be prescribed based on tongue manifestations and TCM syndrome of patients.

ObjectiveTo summarize the characteristics of traditional Chinese medicine （TCM） syndrome in primary Sjögren's syndrome （pSS） patients with interstitial lung disease （ILD） and to explore associated factors. MethodA survey was conducted and pSS patients who were treated in TCM department of rheumatism at China-Japan Friendship Hospital from December 2018 to April 2022 were included. Tongue manifestations and syndromes of patients were recorded. pSS patients with ILD were classified into the pSS-ILD group and those without the ILD were included in the pSS-non-ILD group. The tongue manifestations， syndromes， and laboratory indexes were compared between the two groups， and logistic regression was used to explore the factors associated with pSS-ILD. ResultA total of 200 pSS patients were included， with 186 （93.0%） females， median age of 57 years， and median disease course of 60 months， of which 44 （22%） had pSS-ILD. In terms of tongue manifestations， pSS-ILD patients generally had dark/purple/stasis tongue， fissured tongue， and tongue with little fluid， thick coating， yellow coating， and greasy coating. The proportion patients with yellow coating was higher in pSS-ILD group than in the pSS-non-ILD group （χ²=4.799，P<0.05）. In terms of syndrome， more than 40% of pSS-ILD patients had Qi deficiency， Yin deficiency， phlegm-dampness， Qi stagnation， and/or blood stasis syndrome. As for Yin deficiency， liver-kidney Yin deficiency syndrome ranked the first. For Qi deficiency， lung Qi deficiency syndrome was most commonly seen. The proportion of patients with lung Qi deficiency was higher in the pSS-ILD group than in the pSS-non-ILD group （χ²=18.667，P<0.01）. As to laboratory indexes， compared with the pSS-non-ILD group， pSS-ILD group had high proportion of anti-SSA-positive patients （P<0.05） and high levels of C-reactive protein （CRP） （P<0.01）， complement C3 （χ²=4.332，P<0.05）， and complement C4 （P<0.05）. Logistic regression analysis showed that pSS with ILD was positively associated with lung Qi deficiency ［odds ratio （OR）=6.079， 95% confidence interval （CI） 2.585-14.298， P<0.01）］ and yellow coating （OR=5.260， 95% CI 1.337-20.692， P<0.05） and negatively associated with low C4 （OR=0.199， 95% CI 0.070-0.564， P<0.01）. ConclusionAbout 22% of pSS patients had ILD， and patients with pSS-ILD generally have Qi deficiency， Yin deficiency， phlegm-dampness， Qi stagnation， and/or blood stasis syndrome. Yellow coating， lung Qi deficiency and C4 level are factors associated with pSS combined with ILD.

Objective:To evaluate deep learning for differentiating invasion depth of colorectal adenomas under image enhanced endoscopy (IEE).Methods:A total of 13 246 IEE images from 3 714 lesions acquired from November 2016 to June 2021 were retrospectively collected in Renmin Hospital of Wuhan University, Shenzhen Hospital of Southern Medical University and the First Hospital of Yichang to construct a deep learning model to differentiate submucosal deep invasion and non-submucosal deep invasion lesions of colorectal adenomas. The performance of the deep learning model was validated in an independent test and an external test. The full test was used to compare the diagnostic performance between 5 endoscopists and the deep learning model. A total of 35 videos were collected from January to June 2021 in Renmin Hospital of Wuhan University to validate the diagnostic performance of the endoscopists with the assistance of deep learning model.Results:The accuracy and Youden index of the deep learning model in image test set were 93.08% (821/882) and 0.86, which were better than those of endoscopists [the highest were 91.72% (809/882) and 0.78]. In video test set, the accuracy and Youden index of the model were 97.14% (34/35) and 0.94. With the assistance of the model, the accuracy of endoscopists was significantly improved [the highest was 97.14% (34/35)].Conclusion:The deep learning model obtained in this study could identify submucosal lesions with deep invasion accurately for colorectal adenomas, and could improve the diagnostic accuracy of endoscopists.

Objective:To explore the effectiveness of the artificial intelligence-endoscopic ultrasound (AI-EUS) biliary and pancreatic recognition system in assisting the recognition of EUS images.Methods:Subjects who received EUS due to suspicious biliary and pancreatic diseases from December 2019 to August 2020 were prospectively collected from the database of Department of Gastroenterology, Renmin Hospital of Wuhan University. Pancreatic EUS images of 28 subjects were included for recognition of pancreas standard station. EUS images of bile duct of 29 subjects were included for recognition of bile duct standard station. Eight new endoscopists from the Gastroenterology Department of Renmin Hospital of Wuhan University read the 57 EUS videos with and without the assistance of AI-EUS biliary and pancreatic recognition system. Accuracy of endoscopists' identification of biliary and pancreatic standard sites with and without the assistance of AI-EUS was compared.Results:The accuracy of pancreas standard station identification of the new endoscopists increased from 67.2% (903/1 344) to 78.4% (1 054/1 344) with the assistance of AI-EUS. The accuracy of bile duct standard station identification increased from 56.4% (523/928) to 73.8% (685/928).Conclusion:AI-EUS biliary and pancreatic recognition system can improve the accuracy of EUS images recognition of biliary and pancreatic system, which can assist diagnosis in clinical work.

Objective:To construct a deep learning-based diagnostic system for gastrointestinal submucosal tumor (SMT) under endoscopic ultrasonography (EUS), so as to help endoscopists diagnose SMT.Methods:From January 1, 2019 to December 15, 2021, at the Digestive Endoscopy Center of Renmin Hospital of Wuhan University, 245 patients with SMT confirmed by pathological diagnosis who underwent EUS and endoscopic submucosal dissection were enrolled. A total of 3 400 EUS images were collected. Among the images, 2 722 EUS images were used for training of lesion segmentation model, while 2 209 EUS images were used for training of stromal tumor and leiomyoma classification model; 283 and 191 images were selected as independent test sets to evaluate lesion segmentation model and classification model, respectively. Thirty images were selected as an independent data set for human-machine competition to compare the lesion classification accuracy between lesion classification models and 6 endoscopists. The performance of the segmentation model was evaluated by indexes such as Intersection-over-Union and Dice coefficient. The performance of the classification model was evaluated by accuracy. Chi-square test was used for statistical analysis.Results:The average Intersection-over-Union and Dice coefficient of lesion segmentation model were 0.754 and 0.835, respectively, and the accuracy, recall and F1 score were 95.2%, 98.9% and 97.0%, respectively. Based on the lesion segmentation, the accuracy of classification model increased from 70.2% to 92.1%. The results of human-machine competition showed that the accuracy of classification model in differential diagnosis of stromal tumor and leiomyoma was 86.7% (26/30), which was superior to that of 4 out of the 6 endoscopists(56.7%, 17/30; 56.7%, 17/30; 53.3%, 16/30; 60.0%, 18/30; respectively), and the differences were statistically significant( χ2=7.11, 7.36, 8.10, 6.13; all P<0.05). There was no significant difference between the accuracy of the other 2 endoscopists(76.7%, 23/30; 73.3%, 22/30; respectively) and model(both P<0.05). Conclusion:This system could be used for the auxiliary diagnosis of SMT under ultrasonic endoscope in the future, and to provide a powerful evidence for the selection of subsequent treatment decisions.

Objective:To develop an endoscopic ultrasonography (EUS) station recognition and pancreatic segmentation system based on deep learning and to validate its efficacy.Methods:Data of 269 EUS procedures were retrospectively collected from Renmin Hospital of Wuhan University between December 2016 and December 2019, and were divided into 3 datasets: (1)Dataset A of 205 procedures for model training containing 16 305 images for classification training and 1 953 images for segmentation training; (2)Dataset B of 44 procedures for model testing containing 1 606 images for classification testing and 480 images for segmentation testing; (3) Dataset C of 20 procedures with 150 images for comparing the performance between models and endoscopists. EUS experts (with more than 10 years of experience) A and B classified and labeled all images of dataset A, B and C through discussion, and the results were used as the gold standard. EUS expert C and senior EUS endoscopists (with more than 5 years of experience) D and E classified and labeled the images in dataset C, and the results were used for comparison with model. The main outcomes included accuracy of classification, Dice (F1 score) of segmentation and Cohen Kappa coefficient of consistency analysis.Results:In test dataset B, the model achieved a mean accuracy of 94.1% in classification. The mean Dice of pancreatic and vascular segmentation were 0.826 and 0.841 respectively. In dataset C, the classification accuracy of the model reached 90.0%. The classification accuracy of expert C, senior endoscopist D and E were 89.3%, 88.7% and 87.3%, respectively. The Dice of pancreatic and vascular segmentation in the model were 0.740 and 0.859, 0.708 and 0.778 for expert C, 0.747 and 0.875 for senior endoscopist D, and 0.774 and 0.789 for senior endoscopist E. The model was comparable to the expert level.Consistency analysis showed that there was high consistency between the model and endoscopists (the Kappa coefficient was 0.823 between model and expert C, 0.840 between model and senior endoscopist D, and 0.799 between model and senior endoscopist E).Conclusion:EUS station classification and pancreatic segmentation system based on deep learning can be used for quality control of pancreatic EUS, with a comparable performance of classification and segmentation to that of EUS experts.

Objective: To evaluate the diagnostic efficacy of Japan Narrow Band Imaging Expert Team(JNET) classification under narrow-band imaging (NBI) for colorectal laterally spreading tumors. Methods: Data of 170 laterally spreading tumors (LST) detected by NBI and pigment dyeing were reviewed in the retrospective study. JNET classification under NBI was used for rediagnosis based on surface pattern and vessel pattern. Pit pattern(PP) was observed under pigment dyeing using PP classification. The results were compared with histologic results after endoscopic resection or surgery. Results: The diagnostic sensitivity, specificity, positive predictive value, negative predictive value and accuracy of JNET classification and PP classification were 92.2% VS 70.3%, 82.3% VS 85.0%, 74.7% VS 72.6%, 94.9% VS 83.5%, 85.9% VS 79.7%, respectively (P=0.159). The consistency rates of JNET classification and PP classification in predicting shallow invasion depth of LST were 6.1% and 8.3% respectively and the consistency rates in predicting deep invasion were 30.8% and 4.8%, respectively. Conclusion JNET classification under NBI is effective in predicting malignant laterally spreading tumors, however, its efficacy in predicting tumor invasion depth is unsatisfied. PP classification can be used to improve the diagnostic accuracy for those with diagnostic difficulty.