Objective:To develop a comprehensive training curriculum to enhance the effective implementation of the national initiative promoting dementia care and prevention.Methods:The Delphi method was utilized in an expert consultation that included 44 participants.The initial draft of the training curriculum was developed based on the current state of dementia care and prevention.This draft was subsequently evaluated for its importance, feasibility, and ease of dissemination.Experts offered targeted modifications and additional recommendations.Results:The recovery rate of the expert consultation questionnaire was 95.5%, with a recovery validity rate of 90.9%.The expert authority coefficient was 0.91, and the Kendall's coordination coefficient( W)for expert scoring was 0.316, with a significance level of P<0.001.Four course modules were ultimately identified: the foundation of memory clinic work, the complete management practice skills, group counseling techniques for caregivers, and practical skills for caregivers.The importance of these modules was rated with a mean of 4.92 to 4.95, and the coefficient of variation ranged from 0.044 to 0.063.Each module had a mean value of 4.92 to 4.95, with a coefficient of variation of 0.044 to 0.063; the mean value for practicality was between 4.78 and 4.92, with a coefficient of variation of 0.055 to 0.098; and the mean value for ease of generalization ranged from 4.28 to 4.65, with a coefficient of variation from 0.140 to 0.203.The four modules comprised a total of 55 specific course content items, with the mean value for each item ranging from 4.76 to 5.00 and a coefficient of variation from 0.000 to 0.121.The mean value of usefulness assigned to each entry ranged from 4.55 to 4.98, with a coefficient of variation from 0.031 to 0.150.Additionally, the mean value for ease of propagation assigned to each entry ranged from 4.00 to 4.83, with a coefficient of variation from 0.091 to 0.245. Conclusions:The developed training curriculum, which comprises four course modules and 55 items, demonstrated consistently high levels of importance, practicality, and ease of dissemination.These findings indicate that the curriculum is well-aligned with national initiatives aimed at enhancing dementia care and prevention.

Objective:To develop a deep learning model based on super-resolution endorectal ultrasound（ERUS）images for the preoperative prediction of perineural invasion（PNI）in patients with rectal cancer，thereby providing a reference for risk stratification and individualized treatment planning.Methods:A retrospective analysis was conducted on 382 patients with rectal cancer who underwent total mesorectal excision at Fujian Medical University Union Hospital between June 2019 and February 2024. Patients were randomly divided into a training set（ n=305）and a test set（ n=77）at a ratio of 8∶2，and further grouped into PNI-negative group and PNI-positive group subgroups based on pathological results. Super-resolution ultrasound images were generated from original ERUS images using a generative adversarial network（GAN）. Deep convolutional neural networks were developed based on features from intratumoral and peritumoral regions to identify the optimal region of interest（ROI）. The dSR5_ResNet18 and dSR5_ResNet50 models were constructed using the super-resolution images with a 5-pixel peritumoral extension. Representative clinical features were selected for subgroup analysis based on sample size and intergroup statistical differences between PNI-positive and PNI-negative patients. Forest plots were used to evaluate model applicability and robustness across subgroups. Results:The dSR5_ResNet18 model，built using super-resolution images of the tumor combined with a 5-pixel peritumoral region，achieved the best predictive performance，with an AUC of 0.867（95% CI=0.782 - 0.952）in the test set. Decision curve analysis demonstrated that the dSR5_ResNet18 model provided the greatest net clinical benefit. Forest plot analysis indicated strong generalizability of the models across subgroups such as pathological N stage，maximum lesion length，and lymph node enlargement，though relatively weaker performance was observed in the carcinoembryonic antigen（CEA）subgroup. Among all models，dSR5_ResNet18 exhibited the most consistent performance across subgroups，with the narrowest confidence intervals and highest robustness. Conclusions:The deep learning model incorporating ERUS-based super-resolution reconstruction demonstrated excellent performance in the preoperative prediction of PNI in rectal cancer. It offers significant advantages in image quality and generalizability，and may serve as a valuable tool to assist clinicians in formulating personalized treatment strategies.

Objective To compare and find an optimal model for predicting the risk of DKD occurrence in patients with type 2 diabetes mellitus(T2DM).Methods A total of 2005 patients with T2DM were enrolled in this study from The Second Hospital of Shijiazhuang City during December 2017 to December 2022.All the subjects were divided into a training set(n=1403)and a validation set(n=602)according to the ratio of 3∶1 by simple random sampling.With the occurrence of DKD as the outcome variablein the training set,important feature variables were screened by LASSO regression.Six different machine learning models were established according to the feature variables,thenthe optimal model was determined by comparison,and anonlinerisk predictor for DKD occurrence was constructed in patients with T2DM.Results Taking the occurrence of DKD as the outcome variable in the training set,the results of LASSO regression analysis showed that the optimal value of the model was 10-fold cross validation lambda.1se=0.01662473,and 15 characteristic variables with nonzero coefficient were screened out to be related to the occurrence of DKD.The data included sex,age,family history of DM,DM duration,LDL-C,HbA1c,WBC,PDW,Scr,urine α1-microglobulin,urine β2-microglobulin,urine microalbumin,hypertension,hypokalemia,and DR.In the training set and validation set,the prediction performance of XGBoost model was better than that of other models(AUC=0.872,0.893,95%CI 0.853～0.891,0.865～0.921),the sensitivity was 0.779,0.863,and the specificity was 0.721,0.758,respectively.The F1 scores were 0.774 and 0.787.DCA analysis showed that the XGBoost model had a greater net benefit and threshold probability.According to the XGBoost model,the online predictor of DKD risk in T2DM patients was laid out,and two patients were selected for application,the results showed that the predictive value of the model was 0.185 in non-DKD patients,and the predictive value was 0.510 in DKD patients.Conclusions The XGBoost model is the best model for predicting the occurrence of DKD in T2DM patients,and an online predictor was successfully built.

Objective To compare and find an optimal model for predicting the risk of DKD occurrence in patients with type 2 diabetes mellitus(T2DM).Methods A total of 2005 patients with T2DM were enrolled in this study from The Second Hospital of Shijiazhuang City during December 2017 to December 2022.All the subjects were divided into a training set(n=1403)and a validation set(n=602)according to the ratio of 3∶1 by simple random sampling.With the occurrence of DKD as the outcome variablein the training set,important feature variables were screened by LASSO regression.Six different machine learning models were established according to the feature variables,thenthe optimal model was determined by comparison,and anonlinerisk predictor for DKD occurrence was constructed in patients with T2DM.Results Taking the occurrence of DKD as the outcome variable in the training set,the results of LASSO regression analysis showed that the optimal value of the model was 10-fold cross validation lambda.1se=0.01662473,and 15 characteristic variables with nonzero coefficient were screened out to be related to the occurrence of DKD.The data included sex,age,family history of DM,DM duration,LDL-C,HbA1c,WBC,PDW,Scr,urine α1-microglobulin,urine β2-microglobulin,urine microalbumin,hypertension,hypokalemia,and DR.In the training set and validation set,the prediction performance of XGBoost model was better than that of other models(AUC=0.872,0.893,95%CI 0.853～0.891,0.865～0.921),the sensitivity was 0.779,0.863,and the specificity was 0.721,0.758,respectively.The F1 scores were 0.774 and 0.787.DCA analysis showed that the XGBoost model had a greater net benefit and threshold probability.According to the XGBoost model,the online predictor of DKD risk in T2DM patients was laid out,and two patients were selected for application,the results showed that the predictive value of the model was 0.185 in non-DKD patients,and the predictive value was 0.510 in DKD patients.Conclusions The XGBoost model is the best model for predicting the occurrence of DKD in T2DM patients,and an online predictor was successfully built.

Objective:To develop a comprehensive training curriculum to enhance the effective implementation of the national initiative promoting dementia care and prevention.Methods:The Delphi method was utilized in an expert consultation that included 44 participants.The initial draft of the training curriculum was developed based on the current state of dementia care and prevention.This draft was subsequently evaluated for its importance, feasibility, and ease of dissemination.Experts offered targeted modifications and additional recommendations.Results:The recovery rate of the expert consultation questionnaire was 95.5%, with a recovery validity rate of 90.9%.The expert authority coefficient was 0.91, and the Kendall's coordination coefficient( W)for expert scoring was 0.316, with a significance level of P<0.001.Four course modules were ultimately identified: the foundation of memory clinic work, the complete management practice skills, group counseling techniques for caregivers, and practical skills for caregivers.The importance of these modules was rated with a mean of 4.92 to 4.95, and the coefficient of variation ranged from 0.044 to 0.063.Each module had a mean value of 4.92 to 4.95, with a coefficient of variation of 0.044 to 0.063; the mean value for practicality was between 4.78 and 4.92, with a coefficient of variation of 0.055 to 0.098; and the mean value for ease of generalization ranged from 4.28 to 4.65, with a coefficient of variation from 0.140 to 0.203.The four modules comprised a total of 55 specific course content items, with the mean value for each item ranging from 4.76 to 5.00 and a coefficient of variation from 0.000 to 0.121.The mean value of usefulness assigned to each entry ranged from 4.55 to 4.98, with a coefficient of variation from 0.031 to 0.150.Additionally, the mean value for ease of propagation assigned to each entry ranged from 4.00 to 4.83, with a coefficient of variation from 0.091 to 0.245. Conclusions:The developed training curriculum, which comprises four course modules and 55 items, demonstrated consistently high levels of importance, practicality, and ease of dissemination.These findings indicate that the curriculum is well-aligned with national initiatives aimed at enhancing dementia care and prevention.

Objective:To develop a deep learning model based on super-resolution endorectal ultrasound（ERUS）images for the preoperative prediction of perineural invasion（PNI）in patients with rectal cancer，thereby providing a reference for risk stratification and individualized treatment planning.Methods:A retrospective analysis was conducted on 382 patients with rectal cancer who underwent total mesorectal excision at Fujian Medical University Union Hospital between June 2019 and February 2024. Patients were randomly divided into a training set（ n=305）and a test set（ n=77）at a ratio of 8∶2，and further grouped into PNI-negative group and PNI-positive group subgroups based on pathological results. Super-resolution ultrasound images were generated from original ERUS images using a generative adversarial network（GAN）. Deep convolutional neural networks were developed based on features from intratumoral and peritumoral regions to identify the optimal region of interest（ROI）. The dSR5_ResNet18 and dSR5_ResNet50 models were constructed using the super-resolution images with a 5-pixel peritumoral extension. Representative clinical features were selected for subgroup analysis based on sample size and intergroup statistical differences between PNI-positive and PNI-negative patients. Forest plots were used to evaluate model applicability and robustness across subgroups. Results:The dSR5_ResNet18 model，built using super-resolution images of the tumor combined with a 5-pixel peritumoral region，achieved the best predictive performance，with an AUC of 0.867（95% CI=0.782 - 0.952）in the test set. Decision curve analysis demonstrated that the dSR5_ResNet18 model provided the greatest net clinical benefit. Forest plot analysis indicated strong generalizability of the models across subgroups such as pathological N stage，maximum lesion length，and lymph node enlargement，though relatively weaker performance was observed in the carcinoembryonic antigen（CEA）subgroup. Among all models，dSR5_ResNet18 exhibited the most consistent performance across subgroups，with the narrowest confidence intervals and highest robustness. Conclusions:The deep learning model incorporating ERUS-based super-resolution reconstruction demonstrated excellent performance in the preoperative prediction of PNI in rectal cancer. It offers significant advantages in image quality and generalizability，and may serve as a valuable tool to assist clinicians in formulating personalized treatment strategies.

Humans ; Asthma/drug therapy* ; Biological Therapy

Objective To explore the predictive value of contrast-enhanced echocardiography com-bined with serum levels of CD137 and insulin-like growth factor binding protein 6(IGFBP-6)for cardiovascular adverse events(MACE)in elderly patients with stable coronary heart disease(CHD)after percutaneous coronary intervention(PCI).Methods A total of 108 elderly patients with stable CHD(CHD group)who visited Department of Cardiology of Changsha First Hospital from March 2020 to March 2022 were recruited in this study.They were grouped into a non-MACE group(81 cases)and a MACE group(27 cases)according to whether MACE occurred after PCI.Another 100 healthy individuals who taking physical examination during the same period served as control group.Their serum CD137 and IGFBP-6 levels were detected,and the contrast agent filling speed(β value)and maximum number of microbubbles(A value)were calculated based on the results of contrast-enhanced echocardiography.Their general clinical data were col-lected.ROC curve analysis and multivariate logistic regression analysis were used to analyze the data.Results The serum levels of CD137 and IGFBP-6 were significantly higher,while the β value and A value were obviously lower in the CHD group than the control group(P＜0.01).And the serum levels were notably higher,and the β value and A value were remarkably lower in the MACE group than the non-MACE group(P＜0.01).The AUC of cardiac ultrasound parameters βvalue and A value combined with serum CD137 and IGFBP-6 to predict MACE after PCI in CHD patients was 0.930,which was significantly higher than the AUC value of every single indicator(P＜0.01).β value,A value,CD137 and IGFBP-6 levels were all risk factors for the occurrence of MACE in CHD patients after PCI(P＜0.01).Conclusion Contrast-enhanced echocardiography,serum CD137 and IGFBP-6 levels have certain predictive value for MACE in elderly CHD patients after PCI,and combined detection has higher predictive value.

Objective:To investigate the differences of white matter diffusion properties between vulnerable and resistant individuals to continuous attention after sleep deprivation.Methods:According to the psychomotor vigilance test performance before and after sleep deprivation, the participants were divided into the vulnerable group( n=24) and resistant group( n=25). All participants underwent diffusion tensor imaging (DTI) scans.Tract based spatial statistics(TBSS) was used to compare fractional anisotropy(FA), mean diffusivity(MD), axial diffusivity(AD), radial diffusivity(RD) maps between the two groups.Spearman correlation analysis was conducted by SPSS 24.0 to investigate the relationships between the altered DTI metrics and PVT task performance. Results:(1) Compared with resistant group, FA value of vulnerable group decreased in the body of corpus callosum(x, y, z=-8, 9, 25, t=-7.855), right superior longitudinal fasciculus(x, y, z=-39, -7, 26, t=-6.252), bilateral anterior limb of internal capsule(x, y, z=-13, 8, 13, t=-5.235; x, y, z=12, 8, 3, t=-5.024) and right posterior thalamic radiation(x, y, z=-26, -56, 17, t=-5.469)(TFCE corrected, P<0.05, cluster size≥50 voxel). (2) Compared with resistant group, MD value of vulnerable group increased in the body of corpus callosum(x, y, z=-3, -6, 26, t=7.613), right superior longitudinal fasciculus(x, y, z=-31, -19, 38, t=5.314), bilateral anterior limb of internal capsule(x, y, z=-16, 7, 8, t=6.898; x, y, z=15, 5, 7, t=6.652), splenium of corpus callosum(x, y, z=27, -53, 17, t=6.541), and AD value increased in the right superior longitudinal fasciculus(x, y, z=-33, -19, 39, t=4.892), splenium of corpus callosum(x, y, z=-22, -49, 21, t=5.450), genu of corpus callosum(x, y, z=4, 26, 0, t=4.332), as well as RD value increased in the right superior corona radiata(x, y, z=-17, 1, 33, t=7.558), body of corpus callosum(x, y, z=4, -8, 26, t=6.699), right anterior limb of internal capsule(x, y, z=-12, 7, 3, t=5.212) (TFCE corrected, P<0.05, cluster size≥50 voxel). (3) Correlational analysis revealed that the negative correlations were found between PVT task performance and the FA value in the right superior longitudinal fasciculus( r=-0.492, P<0.001), right anterior limb of internal capsule( r=-0.510, P<0.001), right posterior thalamic radiation( r=-0.502, P<0.001) and body of corpus callosum( r=-0.464, P<0.001). The positive correlations were found between PVT task performance and the MD value in the body of corpus callosum( r=0.500, P<0.001), right superior longitudinal fasciculus( r=0.499, P<0.001), splenium of corpus callosum( r=0.462, P<0.001), right anterior limb of internal capsule( r=0.471, P<0.001), and AD value in right superior longitudinal fasciculus( r=0.643, P<0.001), as well as RD value in right superior corona radiate( r=0.498, P<0.001) (Bonferroni corrected, P<0.003). Conclusion:Differences in the microstructural characteristics of white matter fiber tracts in specific brain regions may constitute the potential neuropathological basis for the phenotypes of vulnerable and resistant individuals to continuous attention after sleep deprivation.

Objective:To investigate the efficacy and safety of fecal microbiota transplantation (FMT) in the treatment of irritable bowel syndrome (IBS), and to explore the effects of FMT on the gut microbiota of IBS patients.Methods:From September 2016 to August 2017, at Guangzhou First People′s Hospital, 28 hospitalized IBS patients who underwent FMT treatment were enrolled. Before FMT, four and 12 weeks after FMT, all the IBS patients completed the irritable bowel syndrome quality of life scale (IBS-QOL), irritable bowel syndrome severity scoring system (IBS-SSS) and gastrointestinal symptom rating scale (GSRS). 16S rDNA sequencing was performed before FMT and four weeks after FMT. The effects of FMT on gut microbiota diversity and microbiota structure of IBS patients were analyzed respectively from the level of phylum, family and genus, and linear discriminant analysis effect size (LEfSe) was further used to screen the different bacteria. Paired t test and paired rank sum test were used for statistical analysis. Results:Twelve weeks after FMT, the scores of the six dimensions of IBS-QOL including dysthymia, behavioral disorder, auto imagery, health concerns, eating avoidance, and relationship expansion were all lower than those before FMT (43.750, 22.656 to 56.250 vs. 48.438, 32.031 to 60.938; 37.500, 18.750 to 56.250 vs. 46.429, 21.429 to 62.500; 31.250, 14.063 to 42.188 vs. 31.250, 18.750 to 50.000; 41.667, 27.083 to 56.250 vs. 50.000, 41.667 to 66.667; 54.167, 43.750 to 72.917 vs. 66.667, 58.333 to 83.333; 8.333, 0.000 to 33.333 vs. 16.667, 8.333 to 33.333, respectively), and the differences were statistically significant ( Z=-2.157, -3.429, -2.274, -3.197, -3.042 and -2.329, all P<0.05). Twelve weeks after FMT, the scores of the two dimensions of IBS-QOL including behavioral disorder and relationship expansion were both lower than those of four weeks after FMT (37.500, 18.750 to 56.250 vs. 39.286, 19.643 to 62.500 and 8.333, 0.000 to 33.333 vs. 16.670, 2.083 to 41.667, respectively), and the differences were statistically significant ( Z=-1.998 and -2.110, both P<0.05). Four and 12 weeks after FMT, the scores of IBS-SSS and GSRS were both lower than those before FMT ((190.32±106.51), (201.43±102.48) vs. (245.93±86.10) and 5.50, 4.00 to 9.00 and 5.50, 4.00 to 8.75 vs. 7.00, 6.00 to 9.75), and the differences were statistically significant ( t=4.402 and 3.848, Z=-3.081 and -3.609; all P<0.01). No serious adverse reactions occurred in the patients after FMT. At the phylum level, after FMT the abundance of Verrucomicrobia in the feces of IBS patients was richer than that before FMT (6.74% vs. 0.37%); at the family level, after FMT the abundance of Verrucomicrobiaceae in the feces of IBS patients was richer than that before FMT (6.74% vs. 0.37%); at the genus level, after FMT the abundance of Akkermansia was richer than that before FMT (6.74% vs. 0.37%); and the differences were statistically significant (all Z=-2.589, all P=0.010). The results of LEfSe method indicated that four weeks after FMT the abundance of Akkermansia in the gut microbiota of IBS patients was richer than that before FMT (6.74% vs. 0.37%), and the difference was statistically significant (linear discriminant analysis value=4.5, P=0.049). Conclusions:FMT is safe and effective in the treatment of IBS. The mechanism may be through upregulating the diversity of gut microbiota and changing the structure of gut microbiota of IBS patients.