Objective:To establish three machine learning prediction models based on CT imaging characteristics of papillary thyroid carcinoma (PTC) , and use SHAP (shapley additive explanations) analysis to investigate the contribution of each CT image features in the best model.Methods:CT imaging features in 426 cases of 440 PTCs confirmed pathologically from Jan. 2016 to Jan. 2021 at the affiliated Hangzhou First People’s Hospital of Westlake University Medical School were retrospectively analyzed. compared with 467 cases of 528 nodular goiter (NG) , evaluating the distribution of four CT characteristics: cookie bite sign, enhanced range of narrowing/blur (ERNB) , microcalcifications, and irregular shape. We split the data into 8∶2 ratio for training and testing sets, then constructed three machine learning models using XGBoost, RF, and SVM. Based on AUC, accuracy, F1 score, and other metrics, we selected the best model. Lastly, we used SHAP values to assess each CT feature’s contribution and positive/negative effects on the model.Results:Among 440 PTC and 528 NG nodules, CT features like cookie bite sign, ERNB, microcalcifications, and irregular shape occurred in 326 and 30 ( χ 2=483.05, P<0.001) , 363 and 106 ( χ 2=374.45, P<0.001) , 158 and 53 ( χ 2=94.24, P<0.001) , and 354 and 52 ( χ 2=491.34, P<0.001) nodules, respectively. The machine learning models built using XGBoost, RF, and SVM had AUC, accuracy, and F1 scores ranging from 0.884~0.925, 0.867~0.873, and 0.844~0.854 respectively on the training set. On the test set, the scores ranged from 0.869~0.923, 0.845~0.871, and 0.803~0.845. Among them, the XGBoost model demonstrated the highest diagnostic performance on the test set. Among the four CT features, irregular shape had the highest absolute SHAP value, positively contributing to PTC diagnosis. Conclusion:XGBoost model showed the highest PTC diagnostic performance. Irregular shape had the greatest positive impact on PTC diagnosis.

Executive function(EF) is an advanced cognitive function of the central nervous system, and is closely related to an individual's capacity for daily living and adaptation. Patients with attention deficit hyperactivity disorder (ADHD) typically exhibit significant executive dysfunction. While most existing studies on the executive function of individuals with ADHD are cross-sectional, and little is known about the longitudinal maturation process of related brain structures and functional connectivity patterns. The findings indicate that ADHD patients exhibit differential developmental trajectories in brain structural and functional connectivity compared with typically developing group.Furthermore, there is a lifespan association between abnormal brain network development and ADHD symptoms. This article aims to elucidate the characteristics of executive function deficits in ADHD patients across different developmental stages, examining their relationship with the nervous system’s development from a development perspective.

Objective To observe the value of non-contrast CT radiomics extreme gradient boosting(XGBoost)model based on SHAP method for predicting acute necrotic collection(ANC)around acute pancreatitis(AP).Methods A total of 307 patients with initially clinically diagnosed AP were retrospectively enrolled.The optimal radiomics features of peripheral pancreatic tissue volume of interest(VOI)were extracted and screened based on automatic segmentation on the first non-contrast CT,and the evaluation results of modified CT severity index(MCTSI)score of AP severity based on first enhanced CT were recorded.The patients were divided into peripancreatic ANC group(ANC group)and acute peripancreatic fluid collection(APFC)group according to follow-up abdominal CT.XGBoost method was used to construct radiomics model,MCTSI model and combined model for predicting AP ANC based on the optimal radiomics features,MCTSI and their combination,respectively.The diagnostic efficacy of each model was evaluated using 5-fold cross-validation method,and the contribution of each variable to combined model was analyzed with SHAP method.Results Among 307 cases,there were 134 cases in ANC group and 173 in APFC group.Totally 6 optimal radiomics features were screened based on the first non-contrast CT.The area under the receiver operating characteristic curve(AUC)of radiomics model,MCTSI model and combined model was 0.936,0.693 and 0.917,respectively.The AUC of MCTSI model was lower than that of radiomics model and combined model(Z=-3.485,-2.824,both P＜0.01),while no significant difference of AUC was found between radiomics model and combined model(Z=-0.817,P=0.415).The contribution of optimal radiomics features to combined model were all higher than that of MCTSI score.Conclusion Non-contrast CT radiomics XGBoost model could effectively predict AP ANC.

This study aims to investigate the medication rules of patented traditional Chinese medi-cine(TCM)compound formulas and molecular mechanisms of core drugs for treating sepsis using data mining and network pharmacology approaches.In the present study,we first searched the PubMed database,Web of Science database,and the China National Knowledge Infrastructure(CNKI)since the establishment of the library to April 30,2024 for the relevant literature on the treatment of sepsis by traditional Chinese medicine.The prescriptions were then statistically ana-lyzed for drug frequency and association analysis to obtain the core drugs.Then we screened the ef-fective active ingredients of the core drugs by TCMSP and other database platforms,obtained sep-sis-related genes in GeneCards and other databases,and statistically intersected targets,and predic-ted the mechanism of action of the core TCMs by subjecting the intersected targets to PPI analy-sis,GO function and KEGG pathway enrichment analysis.Finally,the relationship between key tar-gets and herbal components was examined in reverse by molecular docking method.The results showed that 64 compound formulas were obtained,with a total of 150 Chinese medicines,which were mostly sweet in taste,cold in nature,and belonged to the spleen,stomach and intestinal me-ridians.According to the association rules,the core drugs were identified as"mirabilite-peach ker-nel-rheum officinale".There were 79 intersecting targets between the core drugs and sepsis,with core targets such as IL-1β,EGFR and SRC.MAPK,TNF,IL-17 and other signaling pathways are involved to mediate inflammatory responses,apoptosis and other biological processes to exert ther-apeutic effects on sepsis.The molecular docking results indicated that the docking activity of the key targets with the main components of the drug,and sennoside E_qt has the lowest binding ener-gy and the best docking activity with SRC.In conclusion,this study showed that the prescription of Chinese medicine for sepsis is mostly based on tonifying the spleen and clearing heat.The mecha-nism of action of the core drug"mirabilite-peach kernel-rheum officinale"in the treatment of sep-sis is multilevel and multifaceted,which provides a certain theoretical basis for the treatment of sepsis by traditional Chinese medicine.

BACKGROUND:According to different locations of lumbar disc herniation,it can be classified into many types.Among them,patients with intradural sac type lumbar disc herniation have severe clinical symptoms,which greatly affect their quality of life.Therefore,this article studies the risk factors for its onset,and improving the preoperative diagnostic rate is of great significance for the treatment and prevention of this disease.OBJECTIVE:To explore the risk factors for intradural lumbar disc herniation.METHODS:A retrospective analysis was conducted on 15 patients with intradural lumbar disc herniation(herniation group)admitted to three hospitals in Xuzhou city from May 2014 to November 2022.Propensity score matching was used to match patients with lumbar disc herniation who did not protrude into the dura mater in a ratio of 1:4.A total of 59 patients were selected and included in the non-herniation group.Clinical data were observed in the two groups.Univariate and binary logistic regression analysis,multiple linear regression,and PROBIT regression analysis were utilized to identify the relevant risk factors for intradural lumbar disc herniation.RESULTS AND CONCLUSION:(1)The results of univariate analysis showed that the average disease course,proportion of segments,modified Pfiirmann grading,sacral tilt angle,proportion of previous lumbar spine surgery history,and proportion of heavy manual workers,proportion of lumbar spinal stenosis,redundancy of the cauda equina,and ossification of the posterior longitudinal ligament all showed statistical differences in the herniation and non-herniation groups(P＜0.05).(2)Multivariate regression analysis showed that duration of disease,protruding segment,history of lumbar surgery,modified Pfiirmann grading,cauda equina redundancy,and lumbar spinal stenosis were independent risk factors for intradural lumbar disc herniation.(3)Based on the influencing factors,an receiver operating characteristic curve was constructed,with an area under curve of 0.956(95％CI:0.913-0.998),indicating good discrimination.(4)There was a significant correlation between the duration of the disease,the protruded segment,the history of lumbar spine surgery,the modified Pfirmann classification,cauda equina redundancy,lumbar spinal canal stenosis,and the incidence rate of intradural lumbar disc herniation.

AIM:To investigate the effects of the neutrophil elastase(NE)inhibitor sivelestat on the pathologi-cal changes of cardiac tissues in mice with heart failure with preserved ejection fraction(HFpEF).METHODS:Eight-week-old C57BL/6J mice were randomly divided into 3 groups:control group(n=5),model group(n=6),and treatment group(n=18).The HFpEF mouse model was established using a combined approach of high-fat diet and NG-nitro-L-argi-nine methyl ester(L-NAME).The mice in treatment group received intraperitoneal injection of sivelestat at doses of 12.5,25 and 50 mg·kg-1·d-1(n=6),while those in control and model groups were injected with the same volume of nor-mal saline.After 12 weeks,small animal ultrasound was employed to assess changes in cardiac function,and the lung weight-to-tibia length(LW/TL)ratio was calculated to evaluate pulmonary edema.An exercise fatigue test was conducted to assess exercise intolerance.Histological evaluations were performed using HE,wheat germ agglutinin,dihydroethidium and Sirius red staining to examine overall heart morphology,cross-sectional area of cardiomyocytes,levels of oxidative stress,and the extent of cardiac fibrosis.The expression of fibrosis-related proteins was analyzed using Western blot.RE-SULTS:Compared with model group,sivelestat at various concentrations significantly improved cardiac diastolic function in HFpEF mice,indicated by an increased E/A ratio and a decreased E/e'ratio(P＜0.05).The LW/TL ratio decreased,alleviating lung congestion and exercise intolerance(P＜0.05).The heart weight-to-tibia length(HW/TL)ratio,overall heart cross-sectional area,and cardiomyocyte cross-sectional area all decreased,indicating attenuation in cardiac hypertro-phy(P＜0.05).Additionally,cardiac fibrosis and reactive oxygen species levels were significantly reduced(P＜0.05).CONCLUSION:Inhibition of NE exerts a protective effect against diastolic dysfunction and cardiac fibrosis induced by HFpEF in mice.

Objective:To establish three machine learning prediction models based on CT imaging characteristics of papillary thyroid carcinoma (PTC) , and use SHAP (shapley additive explanations) analysis to investigate the contribution of each CT image features in the best model.Methods:CT imaging features in 426 cases of 440 PTCs confirmed pathologically from Jan. 2016 to Jan. 2021 at the affiliated Hangzhou First People’s Hospital of Westlake University Medical School were retrospectively analyzed. compared with 467 cases of 528 nodular goiter (NG) , evaluating the distribution of four CT characteristics: cookie bite sign, enhanced range of narrowing/blur (ERNB) , microcalcifications, and irregular shape. We split the data into 8∶2 ratio for training and testing sets, then constructed three machine learning models using XGBoost, RF, and SVM. Based on AUC, accuracy, F1 score, and other metrics, we selected the best model. Lastly, we used SHAP values to assess each CT feature’s contribution and positive/negative effects on the model.Results:Among 440 PTC and 528 NG nodules, CT features like cookie bite sign, ERNB, microcalcifications, and irregular shape occurred in 326 and 30 ( χ 2=483.05, P<0.001) , 363 and 106 ( χ 2=374.45, P<0.001) , 158 and 53 ( χ 2=94.24, P<0.001) , and 354 and 52 ( χ 2=491.34, P<0.001) nodules, respectively. The machine learning models built using XGBoost, RF, and SVM had AUC, accuracy, and F1 scores ranging from 0.884~0.925, 0.867~0.873, and 0.844~0.854 respectively on the training set. On the test set, the scores ranged from 0.869~0.923, 0.845~0.871, and 0.803~0.845. Among them, the XGBoost model demonstrated the highest diagnostic performance on the test set. Among the four CT features, irregular shape had the highest absolute SHAP value, positively contributing to PTC diagnosis. Conclusion:XGBoost model showed the highest PTC diagnostic performance. Irregular shape had the greatest positive impact on PTC diagnosis.

Executive function(EF) is an advanced cognitive function of the central nervous system, and is closely related to an individual's capacity for daily living and adaptation. Patients with attention deficit hyperactivity disorder (ADHD) typically exhibit significant executive dysfunction. While most existing studies on the executive function of individuals with ADHD are cross-sectional, and little is known about the longitudinal maturation process of related brain structures and functional connectivity patterns. The findings indicate that ADHD patients exhibit differential developmental trajectories in brain structural and functional connectivity compared with typically developing group.Furthermore, there is a lifespan association between abnormal brain network development and ADHD symptoms. This article aims to elucidate the characteristics of executive function deficits in ADHD patients across different developmental stages, examining their relationship with the nervous system’s development from a development perspective.

Finite element method, as a verified numerical simulation method, has been widely applied in the biomechanical research of clear aligner treatment. The accuracy of its calculation results is closely related to the construction method of the finite element model. This paper takes the finite element modeling process of clear aligner treatment as a clue, and reviews the modeling methods of relevant research in recent years, aiming to provide reference for subsequent studies.

ObjectiveTo study on the different therapeutic effects and potential mechanisms of Rhei Radix et Rhizoma（RH） before and after steaming with wine on constipation model mice. MethodsFifty-four male ICR mice were randomly divided into control group， model group， lactulose group（1.5 mg·kg^-1）， high， medium and low dose groups of RH and RH steaming with wine（PRH）（8， 4， 1 g·kg^-1）. Except for the control group， the constipation model was replicated by gavage of loperamide hydrochloride（6 mg·kg^-1） in the other groups. After 2 weeks of modeling， each administration group was gavaged with the corresponding dose of drug solution， and the control and model groups were given an equal volume of normal saline， 1 time/d for 2 consecutive weeks. After administration， the feces were collected for 16S rRNA sequencing， the levels of gastrin（GAS）， motilin（MTL）， interleukin-6（IL-6）， γ-interferon（IFN-γ） in the colonic tissue were detected by enzyme-linked immunosorbent assay（ELISA）， the histopathological changes of colon were observed by hematoxylin-eosin（HE） staining， flow cytometry was used to detect the proportion changes of CD4⁺， CD8⁺ and regulatory T cell（Treg） in peripheral blood. ResultsCompared with the control group， the model group showed significantly decrease in fecal number in 24 h， fecal quality and fecal water rate（P<0.01）， the colon was seen to have necrotic shedding of mucosal epithelium， localized intestinal glands in the lamina propria were degenerated， necrotic and atrophied， a few lymphocytes were seen to infiltrate in the necrotic area in a scattered manner， the contents of GAS and MTL， the proportions of CD4⁺， CD8⁺ and Treg were significantly reduced（P<0.01）， the contents of IL-6 and IFN-γ were significantly elevated（P<0.05， P<0.01）. Compared with the model group， the fecal number in 24 h， fecal quality and fecal water rate of high-dose groups of RH and PRH were significantly increased（P<0.05， P<0.01）， the pathological damage of the colon was alleviated to varying degrees， the contents of GAS， MTL， IL-6 and IFN-γ were significantly regressed（P<0.05， P<0.01）， and the proportions of CD4⁺ and CD8⁺ were significantly increased（P<0.01）， although the proportion of Treg showed an upward trend， there was no significant difference. In addition， the results of intestinal flora showed that the number of amplicon sequence variant（ASV） and Alpha diversity were decreased in the model group compared with the control group， and there was a significant difference in Beta diversity， with a decrease in the relative abundance of Lactobacillus and an increase in the relative abundances of Bacillus and Helicobacter. Compared with the model group， the ASV number and Alpha diversity were increased in the high-dose groups of RH and PRH， and there was a trend of regression of Beta diversity to the control group， the relative abundance of Lactobacillus increased， and the relative abundances of Bacillus and Helicobacter decreased. ConclusionRH and PRH can improve dysbacteriosis， promote immune system activation， inhibit the release of inflammatory factors for enhancing the gastrointestinal function， which may be one of the potential mechanisms of their therapeutic effect on constipation.