Objective:To screen the optimal machine learning model for predicting the recurrence condition of hepatocellular carcinoma (HCC) at different time points post-surgery, based on the cutoff value of the Ki67 positive proliferation index condition calculated from recurrence-free survival and combined with various clinical features.Methods:retrospective study included initially treated patients with solitary HCC who underwent radical surgery at the Fifth Medical Center of the PLA General Hospital from January 2013 to March 2023. Data included general clinical data, preoperative laboratory parameters, and surgical pathology information about the subjects. The postoperative recurrence status was assessed by querying the medical record system or by telephone follow-up. The Ki67 positive index cutoff value was determined by the X-tile software based on the patient's recurrence-free survival status and time analysis. Survival rates were calculated using the Kaplan-Meier method, and survival curves were plotted. The study population was randomly divided into training and testing groups in a 7:3 ratio using a computer-generated random number method. The minimum redundancy maximum relevance (mRMR) method was used for feature variable selection. Predictive models for postoperative HCC recurrence conditions in patients with HCC were constructed using random forest, support vector machine, logistic regression, and gradient boosting decision tree machine learning algorithms. Inter-group comparisons for continuous data were performed using the t-test or Mann-Whitney U test. Inter-group comparisons of enumeration data were performed using the Pearson χ2 test, continuity-corrected χ2 test, or Fisher's exact test. Results:The cutoff values for the Ki67 positivity index were 0.3 and 0.5 in 510 cases, with a follow-up time ranging from 1.2 to 11.4 years (median: 6.2 years). The recurrence-free survival time was between 1 and 135 months (median: 32 months), with recurrence-free survival rates post-surgery at 1, 2, 3, and 5 years were 87.5%, 77.1%, 61.2%, and 54.5%, respectively. The top five variables predicted HCC recurrence and non-recurrence conditions following surgical follow-up at 6 months, 1 year, 2 years, and beyond 2 years, in accordance with information obtained by the mRMR screen out. The Ki67 positivity index screened a successfully constructed machine learning model to predict HCC recurrence and non-recurrence conditions following surgical follow-up at 6 months, 1 year, 2 years, and beyond 2 years. The machine learning model based on the gradient boosting decision tree algorithm had the best prediction performance among them (areas under the receiver operating characteristic curves for predicting HCC recurrence within six months in the training and validation sets were 0.996 and 0.946, and accuracies were 0.972 and 0.935, respectively).Conclusion:A machine learning model was successfully constructed using the Ki67 positivity index combined with four readily available clinical features to predict HCC recurrence. The machine learning model based on the gradient boosting decision tree algorithm demonstrated the best performance in terms of predicting HCC recurrence within six months after surgery.

OBJECTIVES: This study aims to investigate the optimal dose ratio and mechanisms of the primary active components in Yinchenhao decoction (geniposide, chlorogenic acid, and rhubarb polysaccharides) for ameliorating metabolic-associated steatotic liver disease (MASLD). METHODS: C57BL/6 mice were randomly divided into the normal control group, model control group, uniform design groups 1-6, and Yinchenhao Decoction group; except for the normal control group, mice in all other groups were fed a Western diet to establish a MASLD model, and after 8 weeks of modeling, mice in the uniform design groups 1-6 and Yinchenhao Decoction group were given the corresponding drugs by gavage. At 12 weeks, all mice were sacrificed: their body weight and liver weight were measured, hematoxylin-eosin staining was used to observe the histopathological changes of liver tissue, the plasma levels of alanine transaminase (ALT) and aspartate transaminase (AST) were detected, and the levels of total cholesterol (TC) and triglycerides (TG) in plasma and liver were measured. Based on these results, the optimal uniform design group was identified; subsequently, with plasma AST, plasma TG, and liver TC levels as screening indicators, the optimal dose ratio was obtained via a regression equation, which was further verified from two dimensions, namely functional indicators and tissue morphology. Meanwhile, glucose tolerance test and insulin tolerance test were conducted to evaluate glucose metabolic homeostasis and insulin sensitivity in mice, periodic acid-Schiff staining was used to observe glycogen accumulation, quantitative reverse transcription-polymerase chain reaction was employed to detect the mRNA expression of genes related to glycolipid metabolism and bile acid metabolism, Western blotting was performed to measure the protein expression of molecules involved in bile acid metabolism, and commercial kits were used to determine the plasma levels of total bilirubin (TBIL), direct bilirubin (DBIL), and total bile acid (TBA). RESULTS: Combinations of geniposide, chlorogenic acid, and rhubarb polysaccharide all reduced the liver-to-body weight ratio, alleviated liver injury, and decreased lipid accumulation, among which the uniform design group 6 (200 mg/kg geniposide+160 mg/kg chlorogenic acid+340 mg/kg rhubarb polysaccharide) exhibited the optimal efficacy. Meanwhile, regression analysis indicated that the dosage ratio of uniform design group 6 was the optimal one for MASLD intervention. Validation experiments showed that, compared with single-drug intervention, the optimal dosage ratio resulted in significantly lower body weight, as well as lower plasma levels of ALT, AST and TC in mice (all P<0.05), along with a more pronounced reduction in the area of hepatic lipid accumulation. Mechanistic investigation experiments demonstrated that intervention with the optimal dosage ratio significantly improved glucose tolerance and insulin sensitivity in mice (all P<0.05), reduced hepatic glycogen deposition, and downregulated the mRNA expression of glycolipid metabolism-related genes such as Gsk3, G6pc, Pck1, Fbp1, Fasn, Srebp-1c, Scd1, Slc27a2, and Slc27a5 (all P<0.05); it also decreased plasma levels of TBIL, DBIL, and TBA (all P<0.05), reversed the abnormal protein expression of bile salt export pump (BSEP), farnesoid X receptor (FXR), and cholesterol-7α-hydroxylase (CYP7A1) in the liver (all P<0.05), and reversed the abnormal mRNA expression of bile acid metabolism-related genes including Nr1h4, Cyp7a1, Cyp27a1, Slc10a1, and Slco1a1 (all P<0.05). CONCLUSIONS The combination of geniposide (200 mg/kg), chlorogenic acid (160 mg/kg), and rhubarb polysaccharide (340 mg/kg) exerts the optimal ameliorative effect on MASLD in mice. This superior efficacy is presumably achieved by synergistically regulating the key nodes of glycolipid metabolism and bile acid metabolism, ultimately optimizing the therapeutic outcome.

Objective:To screen the optimal machine learning model for predicting the recurrence condition of hepatocellular carcinoma (HCC) at different time points post-surgery, based on the cutoff value of the Ki67 positive proliferation index condition calculated from recurrence-free survival and combined with various clinical features.Methods:retrospective study included initially treated patients with solitary HCC who underwent radical surgery at the Fifth Medical Center of the PLA General Hospital from January 2013 to March 2023. Data included general clinical data, preoperative laboratory parameters, and surgical pathology information about the subjects. The postoperative recurrence status was assessed by querying the medical record system or by telephone follow-up. The Ki67 positive index cutoff value was determined by the X-tile software based on the patient's recurrence-free survival status and time analysis. Survival rates were calculated using the Kaplan-Meier method, and survival curves were plotted. The study population was randomly divided into training and testing groups in a 7:3 ratio using a computer-generated random number method. The minimum redundancy maximum relevance (mRMR) method was used for feature variable selection. Predictive models for postoperative HCC recurrence conditions in patients with HCC were constructed using random forest, support vector machine, logistic regression, and gradient boosting decision tree machine learning algorithms. Inter-group comparisons for continuous data were performed using the t-test or Mann-Whitney U test. Inter-group comparisons of enumeration data were performed using the Pearson χ2 test, continuity-corrected χ2 test, or Fisher's exact test. Results:The cutoff values for the Ki67 positivity index were 0.3 and 0.5 in 510 cases, with a follow-up time ranging from 1.2 to 11.4 years (median: 6.2 years). The recurrence-free survival time was between 1 and 135 months (median: 32 months), with recurrence-free survival rates post-surgery at 1, 2, 3, and 5 years were 87.5%, 77.1%, 61.2%, and 54.5%, respectively. The top five variables predicted HCC recurrence and non-recurrence conditions following surgical follow-up at 6 months, 1 year, 2 years, and beyond 2 years, in accordance with information obtained by the mRMR screen out. The Ki67 positivity index screened a successfully constructed machine learning model to predict HCC recurrence and non-recurrence conditions following surgical follow-up at 6 months, 1 year, 2 years, and beyond 2 years. The machine learning model based on the gradient boosting decision tree algorithm had the best prediction performance among them (areas under the receiver operating characteristic curves for predicting HCC recurrence within six months in the training and validation sets were 0.996 and 0.946, and accuracies were 0.972 and 0.935, respectively).Conclusion:A machine learning model was successfully constructed using the Ki67 positivity index combined with four readily available clinical features to predict HCC recurrence. The machine learning model based on the gradient boosting decision tree algorithm demonstrated the best performance in terms of predicting HCC recurrence within six months after surgery.

Delirium has a high incidence and is associated with a poor patient prognosis.The pathogenesis of delirium is still unclear.Although antipsychotic drugs are the main intervention in clinical practice,there are currently no drugs that can definitively improve the long-term health-related quality of life in patients with delirium.In this study,we review research on the associations between epigenetic modifications such as non-coding RNAs,DNA methylation,and histone acetylation and the development of delirium,with the aim of providing information to support clinical decisions on the early identification and diagnosis of delirium,as well as its treatment and prognosis and the development of relevant targeted drugs.

Delirium has a high incidence and is associated with a poor patient prognosis.The pathogenesis of delirium is still unclear.Although antipsychotic drugs are the main intervention in clinical practice,there are currently no drugs that can definitively improve the long-term health-related quality of life in patients with delirium.In this study,we review research on the associations between epigenetic modifications such as non-coding RNAs,DNA methylation,and histone acetylation and the development of delirium,with the aim of providing information to support clinical decisions on the early identification and diagnosis of delirium,as well as its treatment and prognosis and the development of relevant targeted drugs.

Background::The goal of the assisted reproductive treatment is to transfer one euploid blastocyst and to help infertile women giving birth one healthy neonate. Some algorithms have been used to assess the ploidy status of embryos derived from couples with normal chromosome, who subjected to preimplantation genetic testing for aneuploidy (PGT-A) treatment. However, it is currently unknown whether artificial intelligence model can be used to assess the euploidy status of blastocyst derived from populations with chromosomal rearrangement.Methods::From February 2020 to May 2021, we collected the whole raw time-lapse videos at multiple focal planes from in vitro cultured embryos, the clinical information of couples, and the comprehensive chromosome screening results of those blastocysts that had received PGT treatment. Initially, we developed a novel deep learning model called the Attentive Multi-Focus Selection Network (AMSNet) to analyze time-lapse videos in real time and predict blastocyst formation. Building upon AMSNet, we integrated additional clinically predictive variables and created a second deep learning model, the Attentive Multi-Focus Video and Clinical Information Fusion Network (AMCFNet), to assess the euploidy status of embryos. The efficacy of the AMCFNet was further tested in embryos with parental chromosomal rearrangements. The receiver operating characteristic curve (ROC) was used to evaluate the superiority of the model. Results::A total of 4112 embryos with complete time-lapse videos were enrolled for the blastocyst formation prediction task, and 1422 qualified blastocysts received PGT-A ( n = 589) or PGT for chromosomal structural rearrangement (PGT-SR, n = 833) were enrolled for the euploidy assessment task in this study. The AMSNet model using seven focal raw time-lapse videos has the best real-time accuracy. The real-time accuracy for AMSNet to predict blastocyst formation reached above 70% on the day 2 of embryo culture, and then increased to 80% on the day 4 of embryo culture. Combing with 4 clinical features of couples, the AUC of AMCFNet with 7 focal points increased to 0.729 in blastocysts derived from couples with chromosomal rearrangement. Conclusion::Integrating seven focal raw time-lapse images of embryos and parental clinical information, AMCFNet model have the capability of assessing euploidy status in blastocysts derived from couples with chromosomal rearrangement.

Objective:To integrate the best evidence of non-drug intervention of urinary incontinence in elderly women and to formulate practical recommendations.Methods:In this systematic review study, using “elderly woman”,“urinary incontinence”,“bladder training”,“pelvic floor muscle training”,“enuresis”,“leakage of urine” as the key words, the 6S evidence resource pyramid model was used to search in British Medical Journal best practice, Uptodate, World Health Organization, Guidelines International Network, National Institute for Health and Care Excellence, Chinese Medical Association, Scottish Intercollegiate Guideline Network, Registered Nurses Association of Ontario, Cochrane Library, The Joanna Briggs Institute (JBI), New Zealand Guidelines Group, Polish Society of Gynecologists and Obstetricians, PubMed, Embase, Medline, Web of Science, SinoMed, China National Knowledge Infrastructure, WanFang Data, etc. The evidence retrieved included evidence-based knowledge base resources, clinical practice guidelines, expert consensus, systematic review, etc. Data were retrieved from January 1, 2017 to May 1, 2022, and collated from May 2, 2022 to May 25, 2022. Two researchers independently evaluated the quality of literature and extracted data using the AGREE Ⅱ and JBI evidence-based health care center assessment tools. The JBI evidence-based health care center′s evidence pre-rating system and evidence recommendation rating system were applied to rank the evidence; and under the guidance of the evidence structure of JBI, the strength of evidence recommendation was determined and the best evidence was extracted and summarized in combination with the study group discussion and expert opinion.Results:A total of 9 articles were retrieved, including 7 guidelines and 2 systematic reviews; and 6 guidelines were classified as Grade A and 1 as grade B; both 2 systematic reviews were rated as Grade A; 84% (27/32) of the items were evaluated as “Yes”. Evidence were summarized as 34 pieces of best evidence from 6 dimensions, including “overall recommendation, evaluation of type and degree of urinary incontinence, lifestyle change, behavioral therapy, prevention of precipitating factors, intervention in special population”; the flow chart of screening, evaluation, special symptoms, life style and behavior therapy was combed, and the practical suggestions were formed.Conclusions:The overall quality of the literature on non-drug intervention of urinary incontinence in elderly women is high, and the level of evidence is high. Early identification of urinary incontinence types and assessment of disease severity, lifestyle changes, avoidance of predisposing factors and behavioral therapy are the key to non-drug treatment of urinary incontinence in those patients.

Objective To delve into the causal relationship between 211 gut microbiota and sepsis employing bidirectional Mendelian randomization(MR).Methods The gut microbiota genome-wide association study(GWAS)data from the Microbiome Genetics Consortium(MiBioGen,n = 18 340)and sepsis GWAS data from the FinnGen(n = 286 146)were harnessed for this study.Initially,single nucleotide polymorphisms(SNP)significantly associated with the relative abundance of 211 gut microbiota taxa were identified as instrumental variables using predefined selection criteria.The primary analytical approach was characterized by the application of inverse variance weighting(IVW),with the effect measure represented by the odds ratio(OR)to assess the results of MR.To ensure precision and reliability,analyses were conducted,including leave-one-out analysis,heterogeneity testing,and tests for pleiotropy at both coherent and incoherent levels.Results The increased risk of sepsis was associated with the elevated abundance of Collinsella[OR = 1.28,95%confidence interval(95%CI)was 1.06-1.56,P = 0.01]and Ruminococcus(OR = 1.19,95%CI was 1.05-1.35,P = 0.005).Furthermore,a protective effect against the development of sepsis was observed in association with the increased abundance of Prevotella(OR = 0.88,95%CI was 0.79-0.97,P = 0.01)and Firmicutes(OR = 0.86,95%CI was 0.75-0.996,P = 0.04).No obvious heterogeneity and irrelevant level pleiotropy were detected.Conclusion Collinsella and Ruminococcus increase the risk of sepsis,while Prevotella and Firmicutes have protective effects against sepsis.

Objective:To investigate the clinical value of modified acute aortic dissection risk score in the early diagnosis of acute aortic dissection (AAD).Methods:The general, clinical, and imaging data of 162 patients who complained of chest and back pain who received treatment between January 2019 and January 2021 in the Department of Emergency, The Second Hospital of Jiaxing, China were collected for this study. The included patients were divided into control (non-AAD, n = 120) and observation (AAD, n = 42) groups according to whether they were diagnosed with AAD. The indexes with statistical significance between the two groups were analyzed using multivariate logistic regression analysis. A score table was established according to the size of OR value. The modified AAD risk score was predicted using the receiver operating curve. Results:Multivariate logistic regression analysis showed that male sex, family history, sudden severe chest and back pain, bilateral blood pressure asymmetry, hypertension, abnormal ultrasound, and D-dimer level were independent risk factors for the diagnosis of AAD (statistical values = 7.84, 6.96, 7.04, 11.38, 7.12, 8.15, 15.07, 9.11, all P < 0.05). Taking the total score of 5 as the prediction standard, the specificity and sensitivity in the prediction of the occurrence of AAD were 84.94% and 95.43%, respectively. The area under the receiver operating curve regarding the modified AAD risk score was 0.909. Conclusion:The modified AAD risk score can be used to conveniently and quickly predict the occurrence of AAD and has a high predictive value. This study is highly innovative and scientific.

Objective To investigate the appropriate age of primary and secondary school students for cardiopulmonary resuscitation (CPR) training.Methods A total of 437 students aged 9-15 years at 3 to 6 grade in the primary schools or 1 to 2 grade in the secondary schools were selected from 2 Yuyao primary and secondary schools by stratified random sampling between March 2017 and January 2018.The numbers of students with the age of 9,10,11,12,13,14 and 15 y were 61,62,66,64,63,63 and 58,respectively.All students received chest compression training provided by Yuyao emergency department People's Hospital according to the 2015 Cardiopulmonary Resuscitation Guidelines.The training included 30 min theoretic teaching and 6 min practice in the simulator.The quality of chest compression performed by students was assessed;the depth,rate,position and retention of chest compression were recorded.Results The mean depth of chest compression in the students aged 9-15 years was 3.8,4.1,4.6,5.1,5.2,5.6 and 5.6 cm,respectively;the accuracy rate was 24.6％(14/61),25.8％ (16/62),50.2％ (33/66),70.5％ (45/64),79.4％(50/63),88.9％(56/63) and 91.4(53/58),respectively.Compared with the students aged 9-11 years,the mean depth of chest compression was significantly increased and accuracy rate was significantly improved in the students aged 12-15 years (Compared with 9-y students,t=-8.936,-9.502,-10.640 and-11.370;x2=35.019,47.599,63.013 and 65.671;compared with 10-y students,t=-6.927,-8.179,-10.70 and-11.047;x2=24.977,35.967,50.916 and 52.727;compared with 1 1-y students,t=-3.095,-4.177,-6.785 and-6.995;x2=5.586,12.114,22.786 and 24.870;all P＜0.05).The mean rate of chest compression was 110-116/min and its accuracy rate was 86.4％-95.2％;the accuracy rate of chest compression position was 90.9％-96.8％ in all students,there were no significant differences among the 7 groups.The mean retention rate of chest compression in the 7 groups was 81.3％(122/150),67.3％(101/150),64.7％ (94/150),48.0％(72/150),48.7％(73/150),33.3％(50/150) and 27.3％(41/150),respectively.Compared with the students aged 9-11 years,the mean retention rate of chest compression was significantly decreased in the students aged 12-15 years (compared with the 9-y students,x2=36.472,35.179,70.64 and 119.92;compared with 10-y students,x2=11.483,10.728,34.682 and 72.150;compared with 11-y students,x2=6.528,5.927,25.855 and 59.11;all P＜0.05).Correlation analysis showed that the depth (r=0.96,0.89,0.91 and 0.86;P＜0.01) and retention rate (r=-0.99,-0.90,-0.93 and-0.86;all P＜0.01) of chest compression were significantly associated with the age,body weight,height and body mass index of students.Conclusion The students with an age of 12 years or more are able to effectively perform chest compression;thus,12 years and above might be the appropriate age for CPR training.