Thyroid nodules are a common endocrine disorder, and their early detection and accurate diagnosis are crucial for the prevention of thyroid cancer. However, the highly heterogeneous morphology and boundaries of thyroid nodules pose significant challenges to their precise identification and classification. Traditional diagnostic approaches rely heavily on physicians' experience, which increases the risk of misdiagnosis and missed diagnoses. With the rapid advancement of computer-aided diagnosis (CAD) technologies, applying deep learning algorithms to the analysis of thyroid nodule ultrasound images has shown great potential. This paper reviews the latest research progress on deep learning-based CAD methods for thyroid nodules, with a focus on their applications in image preprocessing, segmentation and classification. The advantages and limitations of current techniques are analyzed, and potential future directions are discussed. This review aims to highlight the potential of deep learning in thyroid nodule diagnosis and to provide a foundation for selecting feasible pathways for future clinical applications.
Humans ; Thyroid Nodule/diagnostic imaging* ; Deep Learning ; Ultrasonography/methods* ; Diagnosis, Computer-Assisted/methods* ; Algorithms ; Thyroid Neoplasms/diagnostic imaging* ; Image Processing, Computer-Assisted/methods*

As one of the core pathogenesis during treatment with traditional Chinese medicine，liver heat runs through different stages of liver disease. The interpretation of its meaning in different medicine categories（traditional Chinese medicine，Tibetan medicine，Mongolian medicine，Uygur medicine，Dai medicine，Yao medicine，etc.） is not unified， and the phenomena of the same name with different meanings，confusion， and misappropriation emerge. This seriously restricts the inheritance，innovation， and clinical application of traditional Chinese medicine and ethnomedicine. By tracing and analyzing liver heat， it is found that liver heat in traditional Chinese medicine is caused by disordered rest and diet， as well as internal injury due to emotional disorder， which leads to liver dysfunction， Qi stagnation， and heat turning to fire in the liver meridian. The liver heat in Tibetan medicine is caused by the accumulated heat of the liver nature and the evil heat in the liver， which stimulates the toxin of Chiba fever. The liver heat in Mongolian medicine derives from the abnormal diet and rest， making excessive Sheila accumulate in the liver and causing disease. The above etiologies are all related to diet， rest，exogenous evil，emotion，and so on， and the pathogenesis is related to the imbalance of Qi and the metabolic disorder of organs. The clinical symptoms are pain in the liver region，yellow eyes， bitter mouth， fever，digestion，and loss of appetite. The principle of treatment and compatibility of prescription are heat-based， with auxiliary detoxification. Other ethnomedicine， such as Uygur medicine， Dai medicine， Yao Medicine，Miao medicine， and She medicine do not have a clear discussion on liver heat，and their etiology， pathogenesis， treatment，and prescription are not systematic，mostly based on a single drug or proven prescriptions.Through the systematic tracing，mining，induction，analysis， and arrangement of the liver heat based on existing literature information database in China，this paper regarded syndrome as the outline and disease as the goal，clarified the similarities and differences of the pathogenesis of liver heat in traditional Chinese medicine，and determined the relationship between liver heat and liver disease and the status quo of syndrome and treatment.This review provides evidence and reference for clinical prevention and treatment，as well as drug development for liver disease.

Background:Acute pancreatitis(AP)is a common disease of the digestive system,among which severe acute pancreatitis(SAP)has a high mortality rate.Finding more accurate and convenient methods for early recognition of SAP is one of the major challenges in clinical treatment.Aims:To explore the application value of the controlled attenuation parameter(CAP)of ultrasound elastography in predicting SAP.Methods:A retrospective cohort study was conducted involving 135 AP patients admitted to Jiading Branch of Shanghai General Hospital from February to October 2024.Patients were categorized into non-SAP and SAP groups according to the severity of the disease.Clinical data,local complications,laboratory indicators,and CAP were compared between the two groups.Univariate and multivariate Logistic regression analyses were used to identify independent risk factors for SAP.A SAP prediction model based on CAP was constructed according to the identified risk factors and the minimum Akaike information criterion(AIC).ROC curve and Bootstrap method were used to evaluate the efficacy of the prediction model and conduct internal validation,respectively.Results:There were statistically significant differences between the non-SAP group and SAP group in body mass index(BMI),incidence of hyperlipidemia,etiological composition,incidence of pleural and ascitic fluid,length of hospital stay,incidence of peripancreatic effusion,incidence of pancreatic necrosis,white blood cell count(WBC),D-dimer(D-D)level,blood glucose,triglyceride(TG),C-reactive protein(CRP),neutrophil count,procalcitonin(PCT),interleukin-6(IL-6),free triiodothyronine(FT3),and CAP(all P<0.05).Multivariate Logistic regression analysis showed that pancreatic necrosis(OR=13.39,95%CI:3.10-57.94,P<0.001)and CAP(OR=1.01,95%CI:1.01-1.02,P=0.038)were independent risk factors for SAP.The SAP prediction model based on CAP was formulated as:Logit(P)=-5.884+0.010×CAP+2.839×pancreatic necrosis+0.169×D-D+0.132×blood glucose+0.006×CRP.The model showed an area under the curve(AUC)of 0.834 for predicting SAP,which was superior to CAP alone(P<0.05).Internal validation indicated that the prediction model had high stability and accuracy(C-index=0.808).Conclusions:The prediction model constructed based on CAP has good clinical value for predicting SAP,providing a new perspective and tool for early identification and prognostic assessment of AP.

Objective This study aimed to investigate the causes and influencing factors of unplanned reoperations and to propose strategies for reducing their incidence.Methods A retrospective analysis was conducted by comparing medical records of patients with and without unplanned reoperations at a tertiary general hospital in Guangxi throughout 2024.Descriptive statis-tics,chi-square tests,and logistic regression were used to analyze contributing factors.Results Among 43,612 surgeries per-formed in 2024,135 unplanned reoperations were recorded,yielding an overall incidence of 3.09％o.The main causes included postoperative bleeding,complications(e.g.,fistula,infarction,thrombosis),and incision-related issues(e.g.,infection,poor healing).Significant differences(P＜0.05)were observed between the reoperation and non-reoperation groups in gender(x2=11.654),age(x2=19.393),grade of initial surgery(x2=12.405),surgeon's title(x2=14.572),nature of initial surgery(x2=20.287),and patient case classification(x2=21.578).These variables were also identified as significant influencing fac-tors(P＜0.05),with gender,age,nature of surgery,grade Ⅲ surgery,and case classification types B/D being independent risk factors.Conclusion Unplanned reoperation serves as a key indicator of surgical quality and safety.Interventions such as en-hanced hospital-and department-level management,perioperative process optimization,and surgical service mechanism improve-ments are recommended to reduce the incidence of unplanned reoperations.

Objective To explore the optimization strategy of Qishen Granules in treating coronary heart disease with heart failure(CHD-HF)based on data mining and the pathogenic"toxin"theory,and to predict its active components and mechanisms using network pharmacology.Methods Literature on traditional Chinese medicine(TCM)for treating CHD-HF was collected from relevant databases,and prescriptions were screened and established into a database according to inclusion and exclusion criteria.Frequency,association rules,and hierarchical clustering analyses were performed using the Ancient and Modern Medical Case Cloud Platform.Network pharmacology techniques were applied to screen potential targets of the optimized combination for treating CHD-HF,and carry out the targets and pathways enrichment analysis.Results A total of 336 articles and 339 prescriptions involving 191 herbs were included,with 12 herbs used more than 100 times.The core drug combinations for treating CHD-HF included Astragali Radix,Poria,Salviae Miltiorrhizae Radix et Rhizoma,Glycyrrhizae Radix et Rhizoma,Chuanxiong Rhizoma,etc,while commonly used detoxifying herbs included Leonuri Herb,Coptidis Rhizoma,etc.Association rule analysis yielded 10 two-item associations and 17 three-item associations;clustering analysis grouped the data into 5 categories.Based on data mining and the pathogenic"toxin"theory,the combination for treating CHD-HF was optimized to include Astragali Radix,Salviae Miltiorrhizae Radix et Rhizoma,Aconiti Lateralis Radix Praeparata,Glycyrrhizae Radix et Rhizoma,Coptidis Rhizoma,and Taraxaci Herba.Network pharmacology analysis identified 366 common targets between the optimized combination and CHD-HF,with 16 core targets screened out.Kyoto Encyclopedia of Genes and Genomes(KEGG)analysis revealed significant enrichment in pathways such as cancer pathways,lipid and atherosclerosis,Rap1 signaling pathway,hypoxia-inducible factor 1(HIF-1)signaling pathway,phosphatidylinositol-3-kinase/protein kinase B(PI3K/Akt)signaling pathway,Ras signaling pathway,and mitogen-activated protein kinase(MAPK)signaling pathway.Conclusion TCM treatment for CHD-HF primarily focuses on replenishing qi and warming yang,activating blood circulation and resolving fluid retention.Based on data mining results and the pathogenic"toxin"theory,the formulation strategy of Qishen Granules for treating CHD-HF was optimized,potentially exerting therapeutic effects through anti-inflammatory,anti-apoptotic,and anti-hypoxia physiological processes.

Objective This study aimed to investigate the causes and influencing factors of unplanned reoperations and to propose strategies for reducing their incidence.Methods A retrospective analysis was conducted by comparing medical records of patients with and without unplanned reoperations at a tertiary general hospital in Guangxi throughout 2024.Descriptive statis-tics,chi-square tests,and logistic regression were used to analyze contributing factors.Results Among 43,612 surgeries per-formed in 2024,135 unplanned reoperations were recorded,yielding an overall incidence of 3.09％o.The main causes included postoperative bleeding,complications(e.g.,fistula,infarction,thrombosis),and incision-related issues(e.g.,infection,poor healing).Significant differences(P＜0.05)were observed between the reoperation and non-reoperation groups in gender(x2=11.654),age(x2=19.393),grade of initial surgery(x2=12.405),surgeon's title(x2=14.572),nature of initial surgery(x2=20.287),and patient case classification(x2=21.578).These variables were also identified as significant influencing fac-tors(P＜0.05),with gender,age,nature of surgery,grade Ⅲ surgery,and case classification types B/D being independent risk factors.Conclusion Unplanned reoperation serves as a key indicator of surgical quality and safety.Interventions such as en-hanced hospital-and department-level management,perioperative process optimization,and surgical service mechanism improve-ments are recommended to reduce the incidence of unplanned reoperations.

Background:Acute pancreatitis(AP)is a common disease of the digestive system,among which severe acute pancreatitis(SAP)has a high mortality rate.Finding more accurate and convenient methods for early recognition of SAP is one of the major challenges in clinical treatment.Aims:To explore the application value of the controlled attenuation parameter(CAP)of ultrasound elastography in predicting SAP.Methods:A retrospective cohort study was conducted involving 135 AP patients admitted to Jiading Branch of Shanghai General Hospital from February to October 2024.Patients were categorized into non-SAP and SAP groups according to the severity of the disease.Clinical data,local complications,laboratory indicators,and CAP were compared between the two groups.Univariate and multivariate Logistic regression analyses were used to identify independent risk factors for SAP.A SAP prediction model based on CAP was constructed according to the identified risk factors and the minimum Akaike information criterion(AIC).ROC curve and Bootstrap method were used to evaluate the efficacy of the prediction model and conduct internal validation,respectively.Results:There were statistically significant differences between the non-SAP group and SAP group in body mass index(BMI),incidence of hyperlipidemia,etiological composition,incidence of pleural and ascitic fluid,length of hospital stay,incidence of peripancreatic effusion,incidence of pancreatic necrosis,white blood cell count(WBC),D-dimer(D-D)level,blood glucose,triglyceride(TG),C-reactive protein(CRP),neutrophil count,procalcitonin(PCT),interleukin-6(IL-6),free triiodothyronine(FT3),and CAP(all P<0.05).Multivariate Logistic regression analysis showed that pancreatic necrosis(OR=13.39,95%CI:3.10-57.94,P<0.001)and CAP(OR=1.01,95%CI:1.01-1.02,P=0.038)were independent risk factors for SAP.The SAP prediction model based on CAP was formulated as:Logit(P)=-5.884+0.010×CAP+2.839×pancreatic necrosis+0.169×D-D+0.132×blood glucose+0.006×CRP.The model showed an area under the curve(AUC)of 0.834 for predicting SAP,which was superior to CAP alone(P<0.05).Internal validation indicated that the prediction model had high stability and accuracy(C-index=0.808).Conclusions:The prediction model constructed based on CAP has good clinical value for predicting SAP,providing a new perspective and tool for early identification and prognostic assessment of AP.

Objective To explore the value of artificial intelligence federated learning system based on chest X-ray films for pathogen diagnosis of community-acquired pneumonia(CAP)in children.Methods Totally 900 cases of CAP children from 2 hospitals were retrospectively enrolled,including bacterial,viral and mycoplasma CAP(each n=300),and chest posteroanterior X-ray films were collected.Meanwhile,chest posteroanterior X-ray films of 5856 children from the publicly available dataset GWCMCx were collected,including 4273 CAP images and 1583 healthy chest images.All above 6756 images were divided into training set(n=5359)and validation set(n=1397)at the ratio of 8∶2.Then a pathogen diagnosis model of children CAP was established based on attention mechanism.Binary and ternary diagnostic algorithms were designed,and federated deployment training was performed.The efficacy of this system for pathogen diagnosis of children CAP was analyzed and compared with DenseNet model.Results Based on all data,the accuracy of the obtained artificial intelligence federated learning system model for diagnosing children CAP was 97.00%,with the area under the curve(AUC)of 0.990.Based on hospital data,the AUC of this system using single imaging data and clinical-imaging data for pathogen diagnosis of children CAP was 0.858 and 0.836,respectively,both better than that of DenseNet model(0.740,both P＜0.05).Conclusion The artificial intelligence federated learning system based on chest X-ray films could be used for pathogen diagnosis of children CAP.

Objective To investigate the current status of hepatitis B virus (HBV) infection among hospitalized patients aged 1-18 years, as well as the status of immunity after hepatitis B vaccination. Methods Related data were collected from the patients aged 1-18 years who were hospitalized in Henan Provincial People's Hospital from July 2020 to July 2021, including serological markers for hepatitis B (HBsAg, anti-HBs, HBeAg, anti-HBe, and anti-HBc) and hepatitis B vaccination. The epidemiological situation of HBV infection was analyzed, as well as the immune effect after vaccination. The trend chi-square test was used for trend analysis. Results A total of 10 658 hospitalized patients were collected, among whom there were 6 372 male patients (59.79%) and 4 286 female patients (40.21%). In this population, the patients with positive HBsAg accounted for 0.28% (30/10 658), with a relatively high proportion of 0.68% and 0.62%, respectively, in the 17-and 18-year age groups; the patients with positive anti-HBs accounted for 51.82% (5 523/10 658), with a relatively high proportion of > 63% in the 1-4 years age groups, and there was a reduction in the proportion of patients with positive anti-HBs (fluctuating around 40%) in the 5-18 years age groups. With the increase in age, the positive rate of anti-HBs tended to decrease in both male and female patients (male: χ 2 =8.217, P =0.004; female: χ 2 =10.048, P =0.002). Conclusion Based on the data of hospitalized patients, HBV infection in the population aged 1-18 years in Henan Province has the characteristics of low prevalence rate and high immunity, and the reduction in the proportion of patients with positive anti-HBs at more than five years after vaccination should be taken seriously in this region.

OBJECTIVE： To investigate the proportion of 8 kinds of ginsenoside to ginsenoside Rg1 in Panax ginseng and the regularity of growth year in Jilin province， and to provide reference for the identification of growth year. METHODS： The samples of garden ginseng， wild-cultivated ginseng and wild ginseng were collected from different growth years （3-30 years） in Jilin province. The contents of 8 ginsenoside （ginsenoside Rg1， Re， Rf， Rb1， Rc， Rb2， Rb3， Rd） in P. ginseng were determined by HPLC. The contents of saponins as well as the proportion of 8 kinds of ginsenoside to ginsenoside Rg1 were calculated； the relationship of the proportion with growth year was investigated. RESULTS： As the increase of growth year， the proportion of 8 kinds of ginsenosides in garden ginseng to ginsenoside Rg1 as well as that of ginsenoside Re， Rb1， Rc， Rd to ginsenoside Rg1 were decreased gradally （P＜0.001）； the proportion of ginsenoside Re to ginsenoside Rg1 in wild-cultivated ginseng decreased first and then increased（P＜0.001）； the proportion of 8 kinds of ginsenosides to ginsenoside Rg1 as well as the proportion of ginsenoside Re and Rb1 to ginsenoside Rg1 were increased gradually in wild ginseng （P＜0.001）； the proportion of ginsenoside Rf， Rb3 to ginsenoside Rg1 in garden ginseng， wild-cultivated ginseng and wild ginseng had no significant difference（P＞0.05）. CONCLUSIONS： Garden ginseng， wild-cultivated ginseng and wild ginseng contain 8 kinds of ginsenosides. The growth year can be predicted preliminarily according to the proportion of ginsenoside Rg1 and ginsenoside Re， Rb1 to ginsenoside Rg1.