Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Objective: Machine learning (ML) has been reported to have better predictive capability than traditional statistical techniques. The aim of this study was to assess the efficacy of ML algorithms and logistic regression (LR) for predicting depressive symptoms during the COVID-19 pandemic. Methods: Analyses were carried out in a national cross-sectional study involving 21,916 participants. The ML algorithms in this study included random forest (RF), support vector machine (SVM), neural network (NN), and gradient boosting machine (GBM) methods. The performance indices were sensitivity, specificity, accuracy, precision, F1-score, and area under the receiver operating characteristic curve (AUC). Results: LR and NN had the best performance in terms of AUCs. The risk of overfitting was found to be negligible for most ML models except for RF, and GBM obtained the highest sensitivity, specificity, accuracy, precision, and F1-score. Therefore, LR, NN, and GBM models ranked among the best models. Conclusion Compared with ML models, LR model performed comparably to ML models in predicting depressive symptoms and identifying potential risk factors while also exhibiting a lower risk of overfitting.

Here, we report a previously unrecognized syndromic neurodevelopmental disorder associated with biallelic loss-of-function variants in the RBM42 gene. The patient is a 2-year-old female with severe central nervous system (CNS) abnormalities, hypotonia, hearing loss, congenital heart defects, and dysmorphic facial features. Familial whole-exome sequencing (WES) reveals that the patient has two compound heterozygous variants, c.304C>T (p.R102*) and c.1312G>A (p.A438T), in the RBM42 gene which encodes an integral component of splicing complex in the RNA-binding motif protein family. The p.A438T variant is in the RRM domain which impairs RBM42 protein stability in vivo. Additionally, p.A438T disrupts the interaction of RBM42 with hnRNP K, which is the causative gene for Au-Kline syndrome with overlapping disease characteristics seen in the index patient. The human R102* or A438T mutant protein failed to fully rescue the growth defects of RBM42 ortholog knockout ΔFgRbp1 in Fusarium while it was rescued by the wild-type (WT) human RBM42. A mouse model carrying Rbm42 compound heterozygous variants, c.280C>T (p.Q94*) and c.1306_1308delinsACA (p.A436T), demonstrated gross fetal developmental defects and most of the double mutant animals died by E13.5. RNA-seq data confirmed that Rbm42 was involved in neurological and myocardial functions with an essential role in alternative splicing (AS). Overall, we present clinical, genetic, and functional data to demonstrate that defects in RBM42 constitute the underlying etiology of a new neurodevelopmental disease which links the dysregulation of global AS to abnormal embryonic development.
Female ; Animals ; Mice ; Humans ; Child, Preschool ; Intellectual Disability/genetics* ; Heart Defects, Congenital/genetics* ; Facies ; Cleft Palate ; Muscle Hypotonia

Aim To investigate the role of metabolites of eicosapentaenoic acid (EPA) in promoting the transdifferentiation of pancreatic α cells to β cells. Methods Male C57BL/6J mice were injected intraperitoneally with 60 mg/kg streptozocin (STZ) for five consecutive days to establish a type 1 diabetes (T1DM) mouse model. After two weeks, they were randomly divided into model groups and 97% EPA diet intervention group, 75% fish oil (50% EPA +25% DHA) diet intervention group, and random blood glucose was detected every week; after the model expired, the regeneration of pancreatic β cells in mouse pancreas was observed by immunofluorescence staining. The islets of mice (obtained by crossing GCG

BACKGROUND:The lower cervical vertebral pedicle is the main stress site of the posterior column of the spine,which is of great significance for the maintenance of the stability of the human center of gravity and the reduction of shock.At present,there are few reports on the characteristics of the internal bone trabeculae,and the characteristics of the joint site of the vertebral pedicle with the articular process and the vertebral body.It is urgent to understand the fine anatomical structure of the vertebral pedicle and the relationship and function of each part. OBJECTIVE:To observe the microanatomical morphology of the vertebral pedicle by Micro-CT scanning of cervical vertebra specimens,and to measure and analyze the microstructure and morphometric parameters of the bone trabecula in the cervical pedicle under normal conditions to evaluate the safety performance of the cervical spine. METHODS:Micro-CT scanning was performed on 31 sets of cervical vertebrae C3-C7.By checking and reconstructing the areas of interest in the bone trabecular within the vertebral pedicle,the morphological characteristics and distribution direction of the bone trabecular within the cervical pedicle were observed,and the bone microstructure parameters were detected,and the differences in the bone microstructure of the C3-C7 vertebral pedicle were analyzed and compared. RESULTS AND CONCLUSION:(1)The Micro-CT images showed that the honeycomb bone trabeculae of the pedicle of the lower cervical spine presented a complex network of microstructures.The trabeculae near the cortical bone were lamellar and relatively compact,extending forward toward the vertebral body and backward toward the articular process lamina.Abatoid bone trabeculae extended into the medullary cavity and transformed into a network structure,and then into rod-shaped bone trabeculae.The rod-shaped bone trabeculae were sparsely distributed in the medullary cavity.(2)Statistical results of morphological parameters of bone trabeculae showed that bone volume fraction values in C4 and C5 were higher than that in C7(P＜0.05).The bone surface/bone volume value in C7 was higher than that in C3,C4 and C6(P＜0.05).The bone surface density of bone trabeculae in C7 was higher than that in C3,C4,C5 and C6(P＜0.05).Trabecular thickness in C7 was higher than that in C3,C4 and C5(P＜0.05).Bone surface/bone volume and bone surface density of the left pedicle bone trabecular were greater than those on the right side(P＜0.05).(3)The microstructural changes of C3-C7 were summarized,in which the load capacity and stress of the C7 pedicle were poor,and the risk of injury was high in this area.

BACKGROUND:Due to the young age of children,the occipital condyle and foramen magnum are not fully developed,and they are prone to various diseases and injuries in the occipitocervical junction,which requires surgical treatment in severe cases.However,anatomical parameters for the development of the occipital condyle and foramen magnum in children are lacking. OBJECTIVE:To measure the morphological structure of the occipital condyle and foramen magnum by three-dimensional reconstruction technique,and to provide important anatomical parameters for occipitocervical junction lesions,related surgical procedures and forensic identification. METHODS:Imaging data of 389 cases of primitive children and adolescents involved in skull base undergoing spiral CT scanning(247 males and 142 females)aged 1-18 years were collected and divided into 1-3-year-old group,4-6-year-old group,7-9-year-old group,10-12-year-old group,13-15-year-old group,and 16-18-year-old group according to their age.Mimics 16.0 software was used to reconstruct the skull base and measure the length and width of the foramen magnum.A formula was used to calculate the area and index of the foramen magnum.We measured the length,width and height of the occipital condyle,the angle between the long axis and the sagittal axis of the occipital condyle(O-S angle),the included angle between the midpoint of the front and back edges of the foramen magnum and the connection between the back edge of occipital condyle and the intersection point of the foramen magnum(F-O angle),and the included angle between the midpoint of the front and back edges of the foramen magnum and the midpoint of the back wall of the sublingual neural tube(F-H angle).Gender,side and age differences were analyzed among the indicators. RESULTS AND CONCLUSION:(1)In foramen magnum measurement,there was no significant difference between sexes in the index of the foramen magnum(P>0.05),but there were significant differences in length,width and area of the foramen magnum(P<0.05).(2)The O-S angle,F-O angle and F-H angle of the occipitral condyle were not significantly different between genders(P>0.05),but length,width and height of the occipital condyle were significantly different between genders(P<0.05).(3)There were no significant differences in the length of the occipital condyle among different groups(P>0.05),but there were significant differences in the width and height of the occipital condyle,O-S angle,F-O angle and F-H angle among different groups(P<0.05).(4)Length,width and area of the foramen magnum,length,width and height of the occipital condyle showed a wavy increasing trend with the increase of age,while O-S,F-O and F-H angles showed a wavy decreasing trend with the increase of age,while the index of the foramen magnum showed no significant change.(5)In conclusion,there are gender and lateral differences in the morphological indexes of the foramen magnum and the occipital condyle in children.These differences can provide an important reference for clinical surgical approach selection and forensic examination.

Objective:To explore the effectiveness of a hysteroscopic endometrial lesion diagnosis model de-veloped based on deep learning(DL)algorithm combined with gradient-weighted class activation mapping(Grad-CAM)visualization technology.Methods:303 hysteroscopy videos(4781 images)of 291 patients who un-derwent hysteroscopy examination in the Department of Gynecology,Renmin Hospital of Wuhan University from June 1,2021 to December 31,2022 were selected.The dataset was divided into a training set(3703 images)and a test set(1078 images)by weight sampling method.After the training set was used for model learning and train-ing,two model architectures,residual neural network(ResNet18)and efficient neural network(EfficientNet-B0),were selected to verify the model in the test set by five-class and two-class classification tasks,respectively.Tak-ing histopathology as the gold standard,the diagnostic efficacy was evaluated to select the optimal model,and the Grad-CAM layer was embedded in the optimal model to output hysteroscopy images of Grad-CAM.Results:①In the five-class classification tasks,the accuracy of EfficientNet-B0 model(93.23％)was higher than that of Res-Net18 model(84.23％);the area under the curve(AUC)of EfficientNet-B0 model in the diagnosis of five disea-ses,including atypical endometrial hyperplasia,endometrial polyps,endometrial cancer,endometrial atypical hy-perplasia,and submucous myoma,was slightly higher than that of ResNet18 model,and the AUC of both models was almost above 0.980.②In the binary classification task of accuracy and the evaluation of specificity,the two models were similar,both above 93.00％,and the sensitivity of EfficientNet-B0 model(91.14％)was significantly better than that of ResNet18 model(77.22％).③EfficientNet-B0 model combined with Grad-CAM algorithm could identify the abnormal areas in the image.After biopsy and pathological examination,it was confirmed that about 95％of the marked areas in the model's output heatmap were lesion areas.Conclusions:The hysteroscopy di-agnostic model developed by EfficientNet-B0 model combined with Grad-CAM has high diagnostic accuracy,sen-sitivity,and specificity,and has application value in the diagnosis of endometrial lesions.