AIM: To assess the gap between the peak of the base curve to the surface of the cornea, as well as examines the correlation between the thickness of the tear film and the fitting of the orthokeratology lens through optical coherence tomography(OCT), providing an intuitive and quantitative clinical evaluation method for the fitting of the orthokeratology lens.METHODS: Myopia patients who fitted orthokeratology at our hospital from January to December 2023 were included. Examinations, including naked vision, slit lamp, non-contact intraocular pressure, ocular fundus, eye position, corneal diameter, corneal topography, tear film rupture time, optometry, etc., were performed on all patients before fitting. The trial lens parameter was selected according to the examination results, and fluorescein staining was performed to evaluate lens fitting state after patients adapted to wearing glasses. According to the performance of fluorescein staining, the inspected eyes are divided into three groups: lens loose group, lens fitting group, and lens tight group. In addition, the tear film thickness of three groups of subjects was measured by OCT, and the differences between the three groups of data were evaluated.RESULTS: A total of 49 myopic patients(77 eyes)were included. The average sphere degree was -3.10±1.25 D, the average cylinder degree was -0.43(-0.75, 0)D, the average horizontal keratometry(HK)was 42.48±0.81 D, and vertical keratometry(VK)was 42.98(42.25, 43.50)D. There were 21 cases(34 eyes)in the lens fitting group, with 13 cases of bilateral eyes, 8 cases of one eye, 14 cases(22 eyes)in the lens loose group, with 8 cases of bilateral eyes, 6 cases of one eye, and 14 cases(21 eyes)in the lens tight group, with 7 cases of bilateral eyes, 7 cases of one eye. There was no statistical difference in the main basic data of the subjects in the three groups(all P>0.05). OCT showed that the tear film thickness of the lens fitting group, the lens loose group, and the lens tight group was 231.18(219.0, 243.0), 220.41(214.0, 224.3), and 249.00(241.5, 258.0)μm, respectively, and there was statistical significance in the thickness among the three groups(P<0.05).CONCLUSION: OCT can serve as a safe and reliable method for measuring the tear film thickness, which can help evaluate the suitability of orthokeratology and provide a non-invasive, more intuitive, and quantitative evaluation method for the fitting and evaluation of orthokeratology.

Objective To optimize the oxygen therapy regimens for infants with pulmonary diseases during bronchoscopy.Methods A prospective randomized,controlled,and single-center clinical trial was conducted on 42 infants who underwent electronic bronchoscopy from July 2019 to July 2021.These infants were divided into a nasal cannula(NC)group and a modified T-piece resuscitator(TPR)group using a random number table.The lowest intraoperative blood oxygen saturation was recorded as the primary outcome,and intraoperative heart rate and respiratory results were recorded as the secondary outcomes.Results Compared with the NC group,the modified TPR group had a significantly higher level of minimum oxygen saturation during surgery and a significantly lower incidence rate of hypoxemia(P<0.05).In the modified TPR group,there were 6 infants with mild hypoxemia,2 with moderate hypoxemia,and 1 with severe hypoxemia,while in the NC group,there were 3 infants with mild hypoxemia,5 with moderate hypoxemia,and 9 with severe hypoxemia(P<0.05).The modified TPR group had a significantly lower incidence rate of intraoperative respiratory rhythm abnormalities than the NC group(P<0.05),but there was no significant difference in the incidence rate of arrhythmias between the two groups(P>0.05).Conclusions Modified TPR can significantly reduce the risk of hypoxemia in infants with pulmonary diseases during electronic bronchoscopy,and TPR significantly decreases the severity of hypoxemia and the incidence of respiratory rhythm abnormalities compared with traditional NC.

Objective:To investigate the iodine nutrition levels of children aged 8 - 10 and pregnant women in iodine deficient areas and iodine adequate areas of Inner Mongolia Autonomous Region (Inner Mongolia for short), and to provide scientific basis for prevention and treatment of iodine deficiency disorders.Methods:From March to July 2022, iodine nutrition analysis was conducted in iodine deficient areas and iodine adequate areas of 104 banners (counties, cities and districts) in 12 league cities in Inner Mongolia. Each monitoring banner (county, city and district) was divided into five sampling areas by east, west, south, north, and center, with one Sumu (township, street) selected from each area. One primary school was selected from each Sumu (township, street), and 40 non-boarding students aged 8 - 10 were selected from each primary school. One third of the students underwent thyroid ultrasound examination. Twenty pregnant women were selected from each of the 5 Sumus (townships, streets) in each monitoring banner (county, city and district). Household salt samples and random urine samples of children and pregnant women were collected to detect salt iodine and urinary iodine levels.Results:In 2022, the median thyroid volume of children aged 8 - 10 in iodine deficient areas and iodine adequate areas of Inner Mongolia was 2.44 ml, and the goiter rate was 1.89% (123/6 496). There were statistically significant differences in thyroid volume and goiter rate between different league cities ( H = 1 229.05, χ 2 = 34.13, P < 0.001). The coverage rate of iodized salt in 12 league cities was 98.51% (30 628/31 090), the consumption rate of qualified iodized salt was 94.42% (29 355/31 090), and the median salt iodine was 22.80 mg/kg. The median urinary iodine of children ( n = 20 968) was 195.00 μg/L, among which the median urinary iodine in 7 league cities was at the appropriate level of iodine nutrition (100 - 199 μg/L), and the median urinary iodine in 5 league cities was at the level of iodine nutrition exceeding the appropriate level (200 - 299 μg/L). The median urinary iodine of pregnant women ( n = 10 122) was 168.00 μg/L, among which, except for Bayannur (149.18 μg/L), the median urinary iodine in other 11 league cities was at the appropriate level of iodine nutrition (150 - 249 μg/L). Conclusions:The overall iodine nutrition of children and pregnant women in Inner Mongolia is at an appropriate level, but some pregnant women still face the risk of iodine deficiency. In the future, the focus of iodine deficiency disorders prevention and control should be on iodine nutrition monitoring for special needs populations.

;Aim To investigate the molecular mechanism of miR-326 inhibiting breast cancer invasion and metastasis by regulating EphB3 expression. Methods RTFQ-PCR was used to examine the expression of miR-326 in normal breast epithelial cells and breast cancer cells and the transfection efficiency of miR-326 overexpression plasmid. EdU cell proliferation assay and Transwell assay were used to examine the changes in proliferation, migration and invasion ability of different subgroups of cells. Dual luciferase assay was used to verify the presence of binding sites for miR-326 and EphB3. Western blot was used to detect the expression of EphB3 in breast cancer cells after overexpression of miR-326. Results RTFQ-PCR results showed that miR-326 was lowly expressed in breast cancer cells and successfully transfected (P < 0. 05). EdU proliferation assay and Transwell assay results showed that overexpression of miR-326 in breast cancer cells inhibited proliferation, migration and invasive ability (P < 0. 05). The results of dual luciferase assay showed that miR-326 could interact with the 3'-UTR of EphB3 (P < 0. 05). Western blot and Transwell assays showed that miR-326 could negatively regulate EphB3 to inhibit invasive metastasis of breast cancer cells (P < 0. 05). Conclusions MiR-326 acts as a cancer suppressor genes in the development of breast cancer and suppresses the invasion and metastasis of breast cancer cells by regulating the expression of EphB3.

Objective To investigate the effects of microRNA(miR)-432-5p on the proliferation,metasta-sis and radiosensitivity of gastric cancer cells and the targeted regulatory relationship with DEAD-box helicase 41(DDX41)in this process.Methods The expressions of miR-432-5p and DDX41 in different gastric cancer cells were detected by quantitative real-time PCR(qRT-PCR).HGC-27 cells were transfected and divided into NC group(not transfected with plasmid),miR-NC group(transfected with miR-432-5p mimics negative con-trol),miR-432-5p group(transfected with miR-432-5p mimics),miR-432-5p+pcDNA-NC group(transfection miR-432-5p and pcDNA-DDX41 negative control)and miR-432-5p+pcDNA-DDX41 group(transfection miR-432-5p mimics and pcDNA-DDX41).After 48 h,qRT-PCR was used to detect the expression of miR-432-5p and DDX41 in each group of cells,MTT assay was used to detect the proliferation activity of each group of cells,Transwell assay was used to detect the migration ability of each group of cells,colony formation assay and flow cytometry were used to detect the radiosensitivity of each group of cells.The dual luciferase reporter gene assay was used to verify the targeting relationship between miR-432-5p and DDX41.The tumor growth of nude mice was detected,and the expression of DDX41 protein in tumor tissues was detected by Western blot.Results Compared with normal cells,cancer of the stomach cells miR-432-5p levels were decreased(P<0.05),DDX41 mRNA levels were increased(P<0.05).Compared with NC group and miR-NC group,the cell survival rate and DDX41 mRNA level in miR-432-5p group were decreased,the number of migrating cells was decreased(P<0.05),the miR-432-5p level was increased,and the radiosensitivity was enhanced(P<0.05).Overexpression of DDX41 reversed the effects of miR-432-5p on proliferation,migration and radiosensitivity of gastric cancer cells(P<0.05).Compared with NC group and miR-NC group,the volume and weight of tumor tissue and the expression of DDX41 protein in miR-432-5p group were significantly decreased(P<0.05).Con-clusion miR-432-5p through targeted cut DDX41 to inhibit the proliferation and metastasis of gastric cancer cells,enhance cell radiation sensitivity.

Objective: To establish a comprehensive diagnostic classification model of lateral cephalograms based on artificial intelligence (AI) to provide reference for orthodontic diagnosis. Methods: A total of 2 894 lateral cephalograms were collected in Department of Orthodontics, Capital Medical University School of Stomatology from January 2015 to December 2021 to construct a data set, including 1 351 males and 1 543 females with a mean age of (26.4± 7.4) years. Firstly, 2 orthodontists (with 5 and 8 years of orthodontic experience, respectively) performed manual annotation and calculated measurement for primary classification, and then 2 senior orthodontists (with more than 20 years of orthodontic experience) verified the 8 diagnostic classifications including skeletal and dental indices. The data were randomly divided into training, validation, and test sets in the ratio of 7∶2∶1. The open source DenseNet121 was used to construct the model. The performance of the model was evaluated by classification accuracy, precision rate, sensitivity, specificity and area under the curve (AUC). Visualization of model regions of interest through class activation heatmaps. Results: The automatic classification model of lateral cephalograms was successfully established. It took 0.012 s on average to make 8 diagnoses on a lateral cephalogram. The accuracy of 5 classifications was 80%-90%, including sagittal and vertical skeletal facial pattern, mandibular growth, inclination of upper incisors, and protrusion of lower incisors. The acuracy rate of 3 classifications was 70%-80%, including maxillary growth, inclination of lower incisors and protrusion of upper incisors. The average AUC of each classification was ≥0.90. The class activation heat map of successfully classified lateral cephalograms showed that the AI model activation regions were distributed in the relevant structural regions. Conclusions: In this study, an automatic classification model for lateral cephalograms was established based on the DenseNet121 to achieve rapid classification of eight commonly used clinical diagnostic items.
Male ; Female ; Humans ; Young Adult ; Adult ; Artificial Intelligence ; Deep Learning ; Cephalometry ; Maxilla ; Mandible/diagnostic imaging*

Objective: To develop a multi-classification orthodontic image recognition system using the SqueezeNet deep learning model for automatic classification of orthodontic image data. Methods: A total of 35 000 clinical orthodontic images were collected in the Department of Orthodontics, Capital Medical University School of Stomatology, from October to November 2020 and June to July 2021. The images were from 490 orthodontic patients with a male-to-female ratio of 49∶51 and the age range of 4 to 45 years. After data cleaning based on inclusion and exclusion criteria, the final image dataset included 17 453 face images (frontal, smiling, 90° right, 90° left, 45° right, and 45° left), 8 026 intraoral images [frontal occlusion, right occlusion, left occlusion, upper occlusal view (original and flipped), lower occlusal view (original and flipped) and coverage of occlusal relationship], 4 115 X-ray images [lateral skull X-ray from the left side, lateral skull X-ray from the right side, frontal skull X-ray, cone-beam CT (CBCT), and wrist bone X-ray] and 684 other non-orthodontic images. A labeling team composed of orthodontic doctoral students, associate professors, and professors used image labeling tools to classify the orthodontic images into 20 categories, including 6 face image categories, 8 intraoral image categories, 5 X-ray image categories, and other images. The data for each label were randomly divided into training, validation, and testing sets in an 8∶1∶1 ratio using the random function in the Python programming language. The improved SqueezeNet deep learning model was used for training, and 13 000 natural images from the ImageNet open-source dataset were used as additional non-orthodontic images for algorithm optimization of anomaly data processing. A multi-classification orthodontic image recognition system based on deep learning models was constructed. The accuracy of the orthodontic image classification was evaluated using precision, recall, F1 score, and confusion matrix based on the prediction results of the test set. The reliability of the model's image classification judgment logic was verified using the gradient-weighted class activation mapping (Grad-CAM) method to generate heat maps. Results: After data cleaning and labeling, a total of 30 278 orthodontic images were included in the dataset. The test set classification results showed that the precision, recall, and F1 scores of most classification labels were 100%, with only 5 misclassified images out of 3 047, resulting in a system accuracy of 99.84%(3 042/3 047). The precision of anomaly data processing was 100% (10 500/10 500). The heat map showed that the judgment basis of the SqueezeNet deep learning model in the image classification process was basically consistent with that of humans. Conclusions: This study developed a multi-classification orthodontic image recognition system for automatic classification of 20 types of orthodontic images based on the improved SqueezeNet deep learning model. The system exhibitted good accuracy in orthodontic image classification.
Humans ; Male ; Female ; Child, Preschool ; Child ; Adolescent ; Young Adult ; Adult ; Middle Aged ; Deep Learning ; Reproducibility of Results ; Radiography ; Algorithms ; Cone-Beam Computed Tomography

Objective:To study the iodine nutrition status of children in non-high water iodine areas of Inner Mongolia Autonomous Region, analyze the monitoring results of iodine deficiency disorders, so as to provide a basis for further guidance on scientific iodine supplementation for children in Inner Mongolia Autonomous Region.Methods:According to the requirements of the National IDD Monitoring Program and the IDD Monitoring Program of Inner Mongolia in 2021, cluster sampling method was used to select non boarding children aged 8 - 10 from 104 counties (cities, districts) in 12 cities within the jurisdiction of Inner Mongolia Autonomous Region. Random urine samples and household salt samples were collected to detect iodine content. At the same time, 1/3 of the selected children were selected for thyroid B-ultrasound examination to measure thyroid volume.Results:In 2021, a total of 19 968 children aged 8 - 10 in non-high water iodine areas of Inner Mongolia were monitored, and 19 968 urine samples were collected, with a median urine iodine of 199.23 μg/L. There was a statistically significant difference in the frequency distribution of urinary iodine in different cities (χ 2 = 839.51, P < 0.001). The median iodine content of children's household edible salt was 22.64 mg/kg, and the coverage rate of iodized salt, qualified iodized salt coverage rate, and non-iodized salt rate in the entire region was 99.05% (19 778/19 968), 94.98% (18 785/19 778), 94.08% (18 785/19 968) and 0.95% (190/19 968), respectively. Thyroid B-ultrasound were performed in 6553 children, the goiter rate was 1.30% (85/6 553), and there were statistically significant differences in goiter rates among children of different ages and cities (χ 2 = 87.09, P < 0.001; χ 2 = 10.40, P = 0.006) . Conclusions:In 2021, the iodine nutrition of children aged 8 - 10 in non-high water iodine areas of Inner Mongolia is at an appropriate level. However, in the future we should continue to adhere to the monitoring of iodine deficiency disorders, strengthen health education on iodine deficiency disorders, consolidate existing achievements in iodine deficiency disease prevention and control, and achieve the goal of continuous elimination of iodine deficiency disorders.

Objective To evaluate the long-term prognostic value of coronary CT angiography-derived fractional flow reserve(CT-FFR)in elderly patients with coronary heart disease(CHD).Methods A retrospective analysis was performed on 1133 patients with clinically suspected CHD from a prospective observational study based on coronary CTA and CT-FFR at the General Hos-pital of Eastern Theater from April 2018 to March 2019,and 330 elderly CHD patients were even-tually included.According to major adverse cardiovascular events(MACE)occurred or not,295 patients were assigned into non-MACE group and 35 patients into the MACE group.Based on cor-onary CTA data,plaque features were analyzed and CT-FFR values were measured in all lesioned vessels.The relationship of plaque features and CT-FFR with MACE was evaluated by using Cox proportional risk regression model,Kaplan-Meier survival curve,and ROC curve analyses.Results The patients with coronary stenosis(≥50％)or CT-FFR value ≤0.8 had a higher risk of MACE(P＜0.01).Univariate Cox analysis showed that coronary stenosis ≥50％and CT-FFR value 0.8 were risk factors of MACE(P＜0.01).After adjusting confounding factors,multivariate Cox analysis indicated that CT-FFR ≤0.8(HR=17.037,95％CI:5.060-57.358,P=0.000)was only independent predictor for MACE.The risk prediction model based on CT-FFR presented better performance than the model based on coronary CTA stenosis(C-index:0.820 vs 0.696,P=0.000).Conclusion CT-FFR≤0.8 is an important independent predictor for long-term MACE in elderly CHD patients.Clinical risk stratification based on CT-FFR may optimize prognostic man-agement strategies in these patients.

Objective:To develop an automated landmark location system applicable to the case of landmark missing.Methods:Four and eighty-one lateral cephalograms, which contained 240 males and 241 females, with an average age of (24.5±5.6) years, taken from January 2015 to January 2021 in the Department of Orthodontics, Capital Medical University School of Stomatology, and met the inclusion criteria were collected. Five postgraduate orthodontic students were the annotators to manually locate 61 possible landmarks in 481 lateral cephalograms. Two assistant professors in the department as reviewers performed calibration. Two professors as arbitrators, made final decision. Data sets were established (341 were used as training set, 40 as validation set, and 100 as test set). In this paper, an automatic landmarks identification and location model based on convolutional neural networks (CNN), CephaNET, was developed. The model was trained by feeding the original image into the feature extraction module and convolutional pose machine (CPM) module to locate landmarks with high accuracy using deep supervision. Training set was enhanced to 1 684 images by histogram equalization, cropping, and adjustment of brightness. The model was trained to compare the Gaussian heat maps output from the network with the set threshold to identify landmark missing cases. Test set of 100 lateral cephalograms was used to test the accuracy of the model. The evaluation criteria used were success detection rate of missing landmark, mean radial error (MRE) and success detection rate (SDR) in the range of 2.0, 2.5, 3.0, 3.5 and 4.0 mm.Results:The model identified and located 61 commonly used landmarks in 0.13 seconds on average. It had an average accuracy of 93.5% in identifying missing landmarks. The MRE of our testing set was (1.19±0.91) mm. SDR of 2.0, 2.5, 3.0, 3.5 and 4.0 mm were 85.4%, 90.2%, 93.5%, 95.4%, 97.0% respectively.Conclusions:The model proposed in this paper could adapt to the absence of landmark in lateral cephalograms and locate 61 commonly used landmarks with high accuracy to meet the requirements of different cephalometric analysis methods.