Objective To explore a deep learning-based automatic bone age estimation model for elbow joint X-ray images of Chinese Han adolescents and children and evaluate its performance.Methods A total of 943(517 males and 426 females)elbow joint frontal view X-ray images of Chinese Han ado-lescents and children aged 6.00 to<16.00 years were collected from East,South,Central and North-west China.Three experimental schemes were adopted for bone age estimation.Scheme 1:Directly in-put preprocessed images into the regression model;Scheme 2:Train a segmentation network using"key elbow joint bone annotations"as labels,then input segmented images into the regression model;Scheme 3:Train a segmentation network using"full elbow joint bone annotations"as labels,then in-put segmented images into the regression model.For segmentation,the optimal model was selected from U-Net,UNet++and TransUNet.For regression,VGG16,VGG19,InceptionV2,InceptionV3,ResNet34,ResNet50,ResNet101 and DenseNet121 models were selected for bone age estimation.The dataset was randomly split into 80%(754 samples)for training and validation for model fitting and hyperparameter tuning,and 20%(189 samples)as an internal test set to test the performance of the trained model.An additional 104 elbow joint X-ray images from the same demographic and age group were col-lected and used as an external test set.Model performance was evaluated by comparing the mean ab-solute error(MAE),root mean square error(RMSE),accuracies within±0.7 years(P±0.7 years)and±1.0 years(P±1.0 years)between the estimated age and the actual age,and by drawing radar charts,scat-ter plots,and heatmaps.Results When segmented with Scheme 3,the UNet++model achieved good segmentation performance with a segmentation loss of 0.000 4 and an accuracy of 93.8%at a learning rate of 0.000 1.In the internal test set,the DenseNet121 model with Scheme 3 yielded the best results with MAE,P±0.7 years and P±1.0 years being 0.83 years,70.03%,and 84.30%,respectively.In the external test set,the DenseNet121 model with Scheme 3 also performed best,with an average MAE of 0.89 years and an average RMSE of 1.00 years.Conclusion When performing automatic bone age estima-tion using elbow joint X-ray images in Chinese Han adolescents and children,it is recommended to use the UNet++model for segmentation.The DenseNet121 model with Scheme 3 achieves optimal per-formance.Using segmentation networks,especially that trained with annotation areas encompassing the full elbow joint including the distal humerus,proximal radius,and proximal ulna,can improve the ac-curacy of bone age estimation based on elbow joint X-ray images.

Objective To explore a deep learning network model suitable for bone age estimation using shoulder joint X-ray images in Chinese Han adolescents.Methods A retrospective collection of 1 286 shoulder joint X-ray images of Chinese Han adolescents aged 12.0 to<18.0 years(708 males and 578 females)was conducted.Using random sampling,approximately 80%of the samples(1 032 cases)were selected as the training and validation sets for model learning,selection and optimization,and the other 20%samples(254 cases)were used as the test set to evaluate the model's generalization ability.The original single-channel shoulder joint X-ray images and dual-channel inputs combining original images with segmentation labels(manually annotated shoulder joint regions multiplied pixel-by-pixel with original images,followed by segmentation via the U-Net++network to retain only key shoulder joint region information)were respectively input into four network models,namely VGG16,ResNet18,ResNet50 and DenseNet121 for bone age estimation.Additionally,manual bone age estimation was con-ducted on the test set data,and the results were compared with the four network models.The mean absolute error(MAE),root mean square error(RMSE),coefficient of determination(R2),and Pear-son correlation coefficient(PCC)were used as main evaluation indicators.Results In the test set,the bone age estimation results of the four models with dual-channel input of shoulder joint X-ray images outperformed those with single-channel input in all four evaluation indicators.Among them,DenseNet121 with dual-channel input achieved best results with MAE of 0.54 years,RMSE of 0.82 years,R2 of 0.76,and PCC(r)of 0.88.Manual estimation yielded an MAE of 0.82 years,ranking second only to dual-channel DenseNet121.Conclusion The DenseNet121 model with dual-channel input combined with original images and segmentation labels is superior to manual evaluation results,and can effectively estimate the bone age of Chinese Han adolescents.

Hepatic encephalopathy is a clinical syndrome of central nervous system dysfunction caused by liver insufficiency.It severely affects the quality of life of patients and may lead to death.Accurate prediction of the risk of developing hepatic encephalopathy is crucial for early intervention and treatment.In order to identify the risk of hepatic encephalopathy in patients in advance,many studies have been devoted to efforts to develop tools and methods to identify the risk of hepatic encephalopathy as early as possible,so as to develop preventive and early management strategies.Most conventional hepatic encephalopathy risk prediction models currently assess the prob-ability of a patient developing hepatic encephalopathy by analysing factors such as clinical data and biochemical indicators,however,their accuracy,sensitivity and positive predictive value are not high.The application of artificial intelligence to clinical predictive modelling is a very hot and promising area,which can use large amounts of data and complex algorithms to improve the accuracy and efficiency of diagnosis and prognosis.To date,there have been few studies using AI techniques to predict hepatic encephalopathy.Therefore,this paper reviews the research progress of hepatic encephalopathy risk prediction models,and also discusses the prospect of AI application in hepatic encephalopathy risk prediction models.It also points out the challenges and future research directions of AI in HE risk prediction model research in order to promote the development and clinical application of hepatic encephalopathy risk prediction models.

Objective:To assess the early physical growth and development of human immunodeficiency virus-exposed uninfected (HEU) children by longitudinally comparing the differences of growth and development between HEU group and the healthy human immunodeficiency virus-unexposed uninfected (HUU) control group of children aged 0 to 18 months.Methods:A retrospective cohort study was designed.Maternal information of the human immunodeficiency virus (HIV) infected mothers and follow-up information at 0, 1, 3, 6, 9, 12, and 18 months postpartum of their children (born between January 2013 and December 2019 in Chengdu City) were collected from the Information System of Prevention of Mother-to-Child Transmission of Human Immunodeficiency Virus Management. The HUU control group was matched with HEU group by maternal age, gestational age at birth, and infant gender at a ratio of 1∶1. There were 385 children each included in the HEU and HUU groups. Matched samples t-test and the multilevel models were used to compared the physical developmental differences between the two groups. Results:Weight for age Z scores (WAZ) at 0, 3, 6 months of HEU group were -0.72±1.03, -0.09±1.18 and 0.05±1.09, respectively, which were all lower than WAZ of HUU group (-0.21±1.04, 0.42±1.19 and 0.41±1.16, respectively), which were all significantly different ( t=8.41, 7.47 and 5.18, respectively, all P<0.001). Length for age Z scores (LAZ) at 3, 6, 12, 18 months of HEU group were -0.23±1.36, -0.01±1.48, -0.18±1.20 and -0.32±1.13, respectively, which were all lower than LAZ of HUU group (0.24±1.26, 0.30±1.26, 0.07±1.11 and 0.04±1.05, respectively), which were all significantly different ( t=6.14, 4.04, 2.72 and 4.30, respectively, all P<0.01). Weight for length Z scores (WLZ) at 0, 3, 6 months of HEU group were -1.05±1.18, 0.23±1.03 and 0.22±0.95, respectively, which were all lower than WLZ of HUU group (-0.20±0.98, 0.44±1.03 and 0.45±1.00, respectively), which were all significantly different ( t=10.90, 2.95 and 2.96, respectively, all P<0.01). After possible confounding factors were corrected, the WAZ of HEU children at 0, 3, 6 months were still lower than those of HUU children, the LAZ of 3, 6, 12, 18 months were still lower than those of HUU children, and the WLZ of 0, 3, 6 months were still lower than those of HUU children. Conclusions:The differences between HEU and HUU children in Chengdu City mainly occur within six months of age, but the differences of body length persist until 18 months of age.Prenatal exposure to HIV infection affects both fetal and postnatal body growth and development.

Objective: To investigate the clinical presentation and genetic characteristics of malignant infantile osteopetrosis. Methods: This was a retrospective case study. Thirty-seven children with malignant infantile osteopetrosis admitted into Beijing Children's Hospital from January 2013 to September 2022 were enrolled in this study. According to the gene mutations, the patients were divided into the CLCN7 group and the TCIRG1 group. Clinical characteristics, laboratory tests, and prognosis were compared between two groups. Wilcoxon test or Fisher exact test were used in inter-group comparison. The survival rate was estimated with the Kaplan-Meier method and the Log-Rank test was used to compare the difference in survival between groups. Results: Among the 37 cases, there were 22 males and 15 females. The age of diagnosis was 0.5 (0.2, 1.0) year. There were 13 patients (35%) and 24 patients (65%) with mutations in CLCN7 and TCIRGI gene respectively. Patients in the CLCN7 group had an older age of diagnosis than those in the TCIRGI group (1.2 (0.4, 3.6) vs. 0.4 (0.2, 0.6) years, Z=-2.60, P=0.008). The levels of serum phosphorus (1.7 (1.3, 1.8) vs. 1.1 (0.8, 1.6) mmol/L, Z=-2.59, P=0.010), creatine kinase isoenzyme (CK-MB) (457 (143, 610) vs. 56 (37, 82) U/L, Z=-3.38, P=0.001) and the level of neutrophils (14.0 (9.9, 18.1) vs. 9.2 (6.7, 11.1) ×10⁹/L, Z=-2.07, P=0.039) at diagnosis were higher in the CLCN7 group than that in the TCIRG1 group. However, the level of D-dimer in the CLCN7 group was lower than that in the TCIRGI group (2.7 (1.0, 3.1) vs. 6.3 (2.5, 9.7) μg/L, Z=2.83, P=0.005). After hematopoietic stem cell transplantation, there was no significant difference in 5-year overall survival rate between the two groups (92.3%±7.4% vs. 83.3%±7.6%, χ²=0.56, P=0.456). Conclusions: TCIRGI gene mutations are more common in children with osteopetrosis. Children with TCIRGI gene mutations have younger age, lower levels of phosphorus, CK-MB, and neutrophils and higher level of D-dimer at the onset. After hematopoietic stem cell transplantation, patients with CLCN7 or TCIRGI gene mutations have similar prognosis.
Child ; Male ; Female ; Humans ; Osteopetrosis/therapy* ; Retrospective Studies ; Prognosis ; Genes, Recessive ; Phosphorus ; Chloride Channels/genetics* ; Vacuolar Proton-Translocating ATPases/genetics*

A 54-year-old asymptomatic man underwent a video-assisted thoracoscopic left pneumonectomy for squamous-cell carcinoma. During the surgery, a complete left pericardial defect was unexpectedly discovered, but no special intervention was made. The preoperative chest CT was reciewed, which showed the heart extended unusually to the left, but the left pericardial defect was not evident. The operation time was 204 min and the patient was discharged from hospital upon recovery 9 days after the surgery. The pathological result indicated moderately differentiated squamous-cell carcinoma (T2N1M0, stage ⅡB), and metastasis was found in the parabronchial lymph nodes (3/5). The patient did not receive chemotherapy after the surgery, and there was no signs of recurrence 6 months after the surgery. Complete pericardial defects usually do not endanger the lives of patients, and if the patient is asymptomatic, pneumonectomy is feasible.

OBJECTIVES: To reduce the dimension of characteristic information extracted from pelvic CT images by using principal component analysis (PCA) and partial least squares (PLS) methods. To establish a support vector machine (SVM) classification and identification model to identify if there is pelvic injury by the reduced dimension data and evaluate the feasibility of its application. METHODS: Eighty percent of 146 normal and injured pelvic CT images were randomly selected as training set for model fitting, and the remaining 20% was used as testing set to verify the accuracy of the test, respectively. Through CT image input, preprocessing, feature extraction, feature information dimension reduction, feature selection, parameter selection, model establishment and model comparison, a discriminative model of pelvic injury was established. RESULTS: The PLS dimension reduction method was better than the PCA method and the SVM model was better than the naive Bayesian classifier (NBC) model. The accuracy of the modeling set, leave-one-out cross validation and testing set of the SVM classification model based on 12 PLS factors was 100%, 100% and 93.33%, respectively. CONCLUSIONS In the evaluation of pelvic injury, the pelvic injury data mining model based on CT images reaches high accuracy, which lays a foundation for automatic and rapid identification of pelvic injuries.
Algorithms ; Bayes Theorem ; Data Mining ; Least-Squares Analysis ; Support Vector Machine

Child ; Chronic Disease ; Epstein-Barr Virus Infections/complications* ; Familial Primary Pulmonary Hypertension/complications* ; Herpesvirus 4, Human ; Humans ; Persistent Infection ; Pulmonary Arterial Hypertension

Objective To compare the performance of three deep-learning models （VGG19, Inception-V3 and Inception-ResNet-V2） in automatic bone age assessment based on pelvic X-ray radiographs. Methods A total of 962 pelvic X ray radiographs taken from adolescents （481 males, 481 females） aged from 11.0 to 21.0 years in five provinces and cities of China were collected, preprocessed and used as objects of study. Eighty percent of these X ray radiographs were divided into training set and validation set with random sampling method and used for model fitting and hyper-parameters adjustment. Twenty percent were used as test sets, to evaluate the ability of model generalization. The performances of the three models were assessed by comparing the root mean square error （RMSE）, mean absolute error （MAE） and Bland-Altman plots between the model estimates and the chronological ages. Results The mean RMSE and MAE between bone age estimates of the VGG19 model and the chronological ages were 1.29 and 1.02 years, respectively. The mean RMSE and MAE between bone age estimates of the Inception-V3 model and the chronological ages were 1.17 and 0.82 years, respectively. The mean RMSE and MAE between bone age estimates of the Inception-ResNet-V2 model and the chronological ages were 1.11 and 0.84 years, respectively. The Bland-Altman plots showed that the mean value of differences between bone age estimates of Inception-ResNet-V2 model and the chronological ages was the lowest. Conclusion In the automatic bone age assessment of adolescent pelvis, the Inception-ResNet-V2 model performs the best while the Inception-V3 model achieves a similar accuracy as VGG19 model.
Adolescent ; Adult ; Age Determination by Skeleton ; Child ; China ; Female ; Humans ; Male ; Pelvis ; Radiography ; Young Adult

OBJECTIVETo study the influence of cow's milk protein allergy (CMPA) on the diagnosis of functional gastrointestinal diseases (FGID) based on the Rome IV standard in infants and young children.

METHODSA total of 84 children aged 1 month to 3 years who were diagnosed with CMPA were enrolled as the case group, and 84 infants and young children who underwent physical examination and had no CMPA were enrolled as the control group. The pediatricians specializing in gastroenterology asked parents using a questionnaire for the diagnosis of FGID based on the Rome IV standard to assess clinical symptoms and to diagnose FGID.

RESULTSThe case group had a significantly higher incidence rate of a family history of allergies than the control group (P<0.05). In the case group, 38 (45%) met the Rome IV standard for the diagnosis of FGID, while in the control group, 13 (15%) met this standard (P<0.05). According to the Rome IV standard for FGID, the case group had significantly higher diagnostic rates of reflex, functional diarrhea, difficult defecation, and functional constipation than the control group (P<0.05). The children who were diagnosed with FIGD in the control group were given conventional treatment, and those in the case group were asked to avoid the intake of cow's milk protein in addition to the conventional treatment. After 3 months of treatment, the case group had a significantly higher response rate to the treatment than the control group (P<0.05).

CONCLUSIONSIn infants and young children, CMPA has great influence on the diagnosis of FGID based on the Rome IV standard. The possibility of CMPA should be considered during the diagnosis of FGID.