Objective To propose a classification model based on a pseudo-bag strategy and feature adjustment for whole slide imaging in pathology.Methods A pseudo-bag generator was constructed to divide a parent bag into 3 pseudo-bags for increasing the number of training bags.Then,a pseudo-bag learning method based on Nystr?m-based algorithm for approximating self-attention and a selective feature fusion method were employed to process the pseudo-bags.Specifically,the pseudo-bag learning method based on Nystr?m-based algorithm for approximating self-attention reduced computational complexity through an improved multi-head self-attention mechanism while deeply extracting instance features to obtain pseudo-bag classification predictions,thereby enhancing pseudo-bag classification accuracy;and the selective feature fusion method refined pseudo-bag features by filtering and extracting relevant instances.Finally,the model adjusted bag features by extracting confounding factors to avoid interference from irrelevant information and further improve classification accuracy.Results The proposed model was evaluated on two datasets(CAMELYON-16 and TCGA-NSCLC)and compared with 10 other methods,and the results demonstrated that the proposed model achieved the best performance.The proposed method reached an accuracy of 0.943 on the CAMELYON-16 dataset and 0.906 on the TCGA-NSCLC dataset.Conclusion The proposed model can significantly improve the accuracy of whole-slide pathological image classification by effectively mitigating the overfitting and avoiding interference from irrelevant information.

A VLi-net based cell nucleus segmentation method integrating convolutional neural networks(CNN)and Vision Transformer(ViT)is proposed to address the limitation that the U-Net with CNN as its backbone is only proficient in capturing local features and has a restricted receptive field.Firstly,to mitigate challenges such as high cost of data annotation and insufficient annotated data,text annotations are introduced to enhance the network's understanding of image information.Secondly,to improve the segmentation performance of VLi-net,ViT and CNN are combined to fully extract global and local features,with multi-receptive field convolution features incorporating into the ViT structure for effectively mitigating the issues of limited local information interaction and single feature representation in ViT.Finally,an interactive fusion module(ViFusion)is used to efficiently fuse the multi-level features from the CNN and ViT branches.Experimental results show that VLi-net achieves a Dice coefficient of 80.85%and a mean intersection over union(MIoU)of 66.83%on the MoNuSeg dataset,obtains a Dice coefficient of 80.53%and a MIoU of 67.54%on the DSB-2018 dataset,and has a Dice coefficient of 86.87%and a MIoU of 77.44%on the TNBC dataset.These findings confirm that VLi-net outperforms other methods across multiple experimental metrics.

A VLi-net based cell nucleus segmentation method integrating convolutional neural networks(CNN)and Vision Transformer(ViT)is proposed to address the limitation that the U-Net with CNN as its backbone is only proficient in capturing local features and has a restricted receptive field.Firstly,to mitigate challenges such as high cost of data annotation and insufficient annotated data,text annotations are introduced to enhance the network's understanding of image information.Secondly,to improve the segmentation performance of VLi-net,ViT and CNN are combined to fully extract global and local features,with multi-receptive field convolution features incorporating into the ViT structure for effectively mitigating the issues of limited local information interaction and single feature representation in ViT.Finally,an interactive fusion module(ViFusion)is used to efficiently fuse the multi-level features from the CNN and ViT branches.Experimental results show that VLi-net achieves a Dice coefficient of 80.85%and a mean intersection over union(MIoU)of 66.83%on the MoNuSeg dataset,obtains a Dice coefficient of 80.53%and a MIoU of 67.54%on the DSB-2018 dataset,and has a Dice coefficient of 86.87%and a MIoU of 77.44%on the TNBC dataset.These findings confirm that VLi-net outperforms other methods across multiple experimental metrics.

Objective To propose a classification model based on a pseudo-bag strategy and feature adjustment for whole slide imaging in pathology.Methods A pseudo-bag generator was constructed to divide a parent bag into 3 pseudo-bags for increasing the number of training bags.Then,a pseudo-bag learning method based on Nystr?m-based algorithm for approximating self-attention and a selective feature fusion method were employed to process the pseudo-bags.Specifically,the pseudo-bag learning method based on Nystr?m-based algorithm for approximating self-attention reduced computational complexity through an improved multi-head self-attention mechanism while deeply extracting instance features to obtain pseudo-bag classification predictions,thereby enhancing pseudo-bag classification accuracy;and the selective feature fusion method refined pseudo-bag features by filtering and extracting relevant instances.Finally,the model adjusted bag features by extracting confounding factors to avoid interference from irrelevant information and further improve classification accuracy.Results The proposed model was evaluated on two datasets(CAMELYON-16 and TCGA-NSCLC)and compared with 10 other methods,and the results demonstrated that the proposed model achieved the best performance.The proposed method reached an accuracy of 0.943 on the CAMELYON-16 dataset and 0.906 on the TCGA-NSCLC dataset.Conclusion The proposed model can significantly improve the accuracy of whole-slide pathological image classification by effectively mitigating the overfitting and avoiding interference from irrelevant information.

A 62-year-old female patient with systemic vasculitis was treated with glucocorticoid and cyclophosphamide for a long time, and then cyclosporine was added due to poor control of the disease. After more than 4 months of cyclosporine treatment, the patient developed a dark red rash on the right lower limb. Later, skin biopsy was performed and pathological examination and immunohistochemical test results suggested Kaposi′s sarcoma. No abnormality was found in the blood lymphocyte subsets count of the patient, and the immune deficiency diseases were not considered. Kaposi′s sarcoma was considered to be related to the immunosuppressive drugs. Cyclophosphamide and cyclosporine were switched to Leigongteng Duogan (雷公藤多苷) and the dose of glucocorticoid was reduced. Seven days later, the rash in the patient was improved; more than 2 months later, the rash on the patient′s right lower limb subsided and the skin was smooth without swelling.

A 62-year-old female patient with systemic vasculitis was treated with glucocorticoid and cyclophosphamide for a long time, and then cyclosporine was added due to poor control of the disease. After more than 4 months of cyclosporine treatment, the patient developed a dark red rash on the right lower limb. Later, skin biopsy was performed and pathological examination and immunohistochemical test results suggested Kaposi′s sarcoma. No abnormality was found in the blood lymphocyte subsets count of the patient, and the immune deficiency diseases were not considered. Kaposi′s sarcoma was considered to be related to the immunosuppressive drugs. Cyclophosphamide and cyclosporine were switched to Leigongteng Duogan (雷公藤多苷) and the dose of glucocorticoid was reduced. Seven days later, the rash in the patient was improved; more than 2 months later, the rash on the patient′s right lower limb subsided and the skin was smooth without swelling.

Objective:To explore the effect of life-style intervention on non-obese polycystic ovary syndrome (PCOS) patients.Methods:This was a prospective self-control study. Forty-three cases of non-obese [18.5 kg/m 2≤body mass index （BMI）<24.0 kg/m 2] PCOS patients were enrolled from May 2017 to March 2018 in Obstetrics and Gynecology Hospital of Fudan University. The individualized dietary and exercise lifestyle intervention was formulated based on the body composition. After 12 weeks, the correlations among body composition data, insulin resistance, reproductive hormone improvement， and ovulation recovery were analyzed. Results:Before intervention, 21 of 43 non-obese PCOS patients had normal body fat rate, 22 patients (51.2%) had higher body fat rate than normal, 17 patients had normal standard proportion of skeletal muscle weight (90%-110%), and 26 patients (60.5%) had lower standard proportion of skeletal muscle weight. Correlation analysis showed that fasting insulin (FINS) and homeostasis model assessment-insulin resistance (HOMA-IR) were positively correlated with body fat rate ( r=0.442, P=0.003; r=0.395, P=0.010, respectively), and negatively correlated with the standard proportion of skeletal muscle weight ( r=-0.492, P=0.001; r=-0.536, P<0.001, respectively). After 12 weeks of lifestyle intervention, the body fat rate of non-obese PCOS patients decreased significantly ( P<0.01); standard proportion of skeletal muscle weight and skeletal muscle weight increased significantly ( P<0.001); HOMA-IR decreased significantly ( P<0.001); level of luteinizing hormone (LH) and decreased significantly （ P=0.005， P=0.004）; and 31 (70%) patients recovered with spontaneous ovulation ( P<0.001）. Correlation analysis showed that: 1) after treatment, standard proportion of skeletal muscle weight in non-obese PCOS patients was negatively correlated with HOMA-IR ( r=-0.512, P=0.001), and insulin-under-curve-area ( r=-0.421, P=0.007); 2)after treatment, the body fat rate was positively correlated with LH and total testerone ( r=0.455, P=0.003; r=0.377, P=0.015); after treatment, standard proportion of skeletal muscle weight was negatively correlated with total testerone ( r=-0.307, P=0.048); 3) after treatment, the LH was positively correlated with HOMA-IR ( r=0.39, P=0.011); 4) regression analysis showed that the recovery of spontaneous ovulation was closely related to the decrease of body weight, the increase of skeletal muscle standard proportion and the decrease of area under insulin curve ( P=0.016, P=0.004, P=0.003). Conclusion:Insulin resistance in non-obese PCOS patients is closely related to the increase of body fat rate and the decrease of the standard proportion of skeletal muscle weight. Individualized lifestyle interventions based on body composition results (focused on increasement of skeletal muscles) can increase insulin sensitivity, reduce LH and testosterone levels in non-obese PCOS patients, and promote the recovery of spontaneous ovulation.

Objective:To explore the effect of life-style intervention on non-obese polycystic ovary syndrome (PCOS) patients.Methods:This was a prospective self-control study. Forty-three cases of non-obese [18.5 kg/m 2≤body mass index （BMI）<24.0 kg/m 2] PCOS patients were enrolled from May 2017 to March 2018 in Obstetrics and Gynecology Hospital of Fudan University. The individualized dietary and exercise lifestyle intervention was formulated based on the body composition. After 12 weeks, the correlations among body composition data, insulin resistance, reproductive hormone improvement， and ovulation recovery were analyzed. Results:Before intervention, 21 of 43 non-obese PCOS patients had normal body fat rate, 22 patients (51.2%) had higher body fat rate than normal, 17 patients had normal standard proportion of skeletal muscle weight (90%-110%), and 26 patients (60.5%) had lower standard proportion of skeletal muscle weight. Correlation analysis showed that fasting insulin (FINS) and homeostasis model assessment-insulin resistance (HOMA-IR) were positively correlated with body fat rate ( r=0.442, P=0.003; r=0.395, P=0.010, respectively), and negatively correlated with the standard proportion of skeletal muscle weight ( r=-0.492, P=0.001; r=-0.536, P<0.001, respectively). After 12 weeks of lifestyle intervention, the body fat rate of non-obese PCOS patients decreased significantly ( P<0.01); standard proportion of skeletal muscle weight and skeletal muscle weight increased significantly ( P<0.001); HOMA-IR decreased significantly ( P<0.001); level of luteinizing hormone (LH) and decreased significantly （ P=0.005， P=0.004）; and 31 (70%) patients recovered with spontaneous ovulation ( P<0.001）. Correlation analysis showed that: 1) after treatment, standard proportion of skeletal muscle weight in non-obese PCOS patients was negatively correlated with HOMA-IR ( r=-0.512, P=0.001), and insulin-under-curve-area ( r=-0.421, P=0.007); 2)after treatment, the body fat rate was positively correlated with LH and total testerone ( r=0.455, P=0.003; r=0.377, P=0.015); after treatment, standard proportion of skeletal muscle weight was negatively correlated with total testerone ( r=-0.307, P=0.048); 3) after treatment, the LH was positively correlated with HOMA-IR ( r=0.39, P=0.011); 4) regression analysis showed that the recovery of spontaneous ovulation was closely related to the decrease of body weight, the increase of skeletal muscle standard proportion and the decrease of area under insulin curve ( P=0.016, P=0.004, P=0.003). Conclusion:Insulin resistance in non-obese PCOS patients is closely related to the increase of body fat rate and the decrease of the standard proportion of skeletal muscle weight. Individualized lifestyle interventions based on body composition results (focused on increasement of skeletal muscles) can increase insulin sensitivity, reduce LH and testosterone levels in non-obese PCOS patients, and promote the recovery of spontaneous ovulation.

Objective:To revise the Chinese version of Devereux Early Childhood Assessment for Preschool-ers,Second Edition (DECA-P2 ),and assess its validity and reliability.Methods:Totally 608 children aged 3 -5 years were rated by their teachers with the Chinese version of DECA-P2 and two criterion scales,which included the Chinese version of Preschool Behavioral and Emotional Rating Scale,and Social Competence and Behavior Evalua-tion Scale-30.Data of 570 were valid.After identifying valid items using item analysis,exploratory factor analysis was performed for half of all samples.Confirmatory factor analysis was also conducted with another half samples.Moreover,the present study tested criterion-related validity,internal consistency reliability and split-half re-liability of the Chinese version of DECA-P2.Results:Through the exploratory factor analysis,the obtained structure of protective factors was the same as the original scale,accounting for 59.82%of the variance.The results of con-firmatory factor analysis achieved the standard of psychometrics (χ2/df=2.50,RMESA=0.07,CFI=0.91,IFI=0.91).Generally,each subscale had significant correlation with the two criterion scales (r=-0.21 ~0.80). Be-sides,the Cronbach αcoefficients of all subscales were between 0.80 and 0.93,and the split-half reliabilities of them were between 0.79 and 0.89.Conclusion:The Chinese version of DECA-P2 has acceptable psychometric quality and can be applied to evaluate resilience for children aged from 3 to 5 years in China.