Adverse drug reactions （ADRs） refer to harmful or unintended reactions unrelated to the intended purpose of medication administration， which can lead to various issues such as accelerated disease progression and prolonged hospitalization. Traditional ADRs monitoring systems （such as spontaneous reporting systems） suffer from limitations such as low reporting rates and inconsistent data quality， which hinder the early prevention and control of ADRs. With the rapid development of information technology， machine learning has emerged as a powerful tool for management and decision-making of ADRs by leveraging its strengths in feature extraction and dynamic temporal pattern analysis. By reviewing relevant literature at home and abroad in recent years， this paper summarizes the progress in the application of machine learning for ADRs prediction. It is found that machine learning has gradually been applied to the early warning and risk prediction of ADRs in target organs such as the kidneys， liver， heart and bone marrow （such as acute kidney injury， drug-induced liver injury， and so on）. Although machine learning demonstrates significant application potential in the field of ADRs prediction， it still faces limitations such as inadequate quality control of clinical data， lack of standardized criteria for model performance evaluation， insufficient model interpretability and difficulties in clinical translation. In the future， the development trend of machine learning in the field of ADRs prediction should follow a “technology-validation-integration” pathway to systematically promote the practical implementation of models.

Objective: To characterize perioperative dynamic changes in immune-cell phenotypes and inflammatory cytokines in patients undergoing CPB (cardiopulmonary bypass) cardiac surgery, and to explore their associations with postoperative outcomes. Methods: In this prospective cohort study, 120 adult patients who underwent elective cardiac surgery under CPB at West China Hospital from May 2022 to March 2023 were enrolled. Perioperative immune-cell phenotypes and concentrations of 40 inflammation-related cytokines were measured. The primary outcomes were the sequential organ failure assessment (SOFA) score at 24 h after surgery and ΔSOFA (the peak SOFA score within 48 h after surgery minus the preoperative SOFA score). Secondary outcomes included major adverse cardiovascular events (MACE), acute kidney injury (AKI), respiratory failure, severe liver injury, and infection. Results: The mean age of enrolled patients was 57±10 years. Of these, 52% (62/120) were male and 90% (108/120) underwent valve surgery. During the rewarming to the end of CPB, neutrophil counts rapidly increased (7.39×10 /L vs preoperative 3.07×10 /L, P<0.001), with significant upregulation of CD11b (7.30×10 /L vs preoperative 3.05×10 /L, P<0.001) and CD54 (7.15×10 /L vs preoperative 2.99×10 /L, P<0.001). Lymphocyte counts increased at the end of CPB (1.75×10 /L vs preoperative 1.12×10 /L, P<0.001) but decreased significantly at 24 h after surgery (0.59×10 /L vs preoperative 1.12×10 /L, P<0.001). Plasma analysis showed that multiple pro-inflammatory cytokines increased during CPB and remained elevated up to 24 h after surgery; five chemokines and the anti-inflammatory cytokine IL-10 peaked at the end of CPB. The SOFA score increased from 1 (1, 2) preoperatively to 7 (5, 10) at 24 h after surgery, with a ΔSOFA of 6 (4, 8). Within 30 days after surgery, 48 patients (40.0%) developed AKI, 17 (14.2%) developed infection, 4 (3.3%) developed severe liver injury, 3 (2.5%) developed respiratory failure, and 3 (2.5%) experienced MACE. During the 2-year follow-up, 8 patients (6.7%) experienced MACE and 5 (4.2%) died. Conclusion: Multi-organ dysfunction is common after cardiac surgery under CPB (median ΔSOFA, 6), accompanied by perioperative activation of multiple immune-cell subsets and upregulation of pro-inflammatory, anti-inflammatory, and chemotactic mediators. This study provides data-driven evidence and research clues for further investigation of the associations between CPB-related immune perturbations and postoperative organ dysfunction and clinical outcomes.

Objective To quantitatively assess the exposure-response association between exposure to heatwave and sudden death, estimate the attributable excess deaths, and identify potential vulnerable subgroups. Methods A time-stratified case-crossover study was conducted among residents who died from sudden death in Jiangsu Province, China between 2015 and 2021. Heatwave events in Jiangsu Province, defined using varying relative temperature thresholds and durations, were identified using temperature data from the China Meteorological Administration Land Data Assimilation System (CLDAS V2.0). Individual heatwave exposure was assessed based on each subject's residential address. The exposure-response association between heatwave and sudden death was evaluated using conditional logistic regression model combined with a Distributed Lag Nonlinear Model(DLNM). Heatwave-attributable excess deaths were estimated. Stratified analyses by sex and age were performed to assess potential effect modifications. Results Under all definitions, exposure to heatwave was significantly associated with an increased risk of sudden death, and the risk increased with the intensity of heatwave. Using the P95_3d definition (temperature exceeding the 95th percentile for ≥3 consecutive days), heatwave was significantlyassociated with a 56% increased risk of sudden death (95% CI: 31%, 86%). The population-attributable fraction of sudden death due to heatwave exposure was 1.45% (95% CI: 0.97%, 1.90%). Stratified analyses indicated no statistically significant differences in the association between heatwave exposure and sudden death across age or sex subgroups. Conclusion Heatwave exposure was associated with an increased risk of sudden death. Reducing heatwave exposure during summer may help lower the occurrence of sudden death.

Objective: To explore the network structure of middle school students basic psychological needs and psychological behavioral problems, and identify the core nodes within the network, as well as examine demographic subgroup differences, so as to provide support for targeted mental health interventions for adolescents. Methods: In September and October of 2023, a total of 2 000 junior and senior high school students were selected with multistage cluster random sampling from 8 schools in Jiaojiang District and Tiantai County, Taizhou City. An online self administered questionnaire was used to assess emotional and behavioral problems, perceived autonomy, self awareness, loneliness, and social support. The instruments included the Strengths and Difficulties Questionnaire (SDQ), Perceived Choice and Awareness of Self Scale (PCASS), Mental Health Literacy Questionnaire (MHLQ), University of California,Los Angeles Loneliness Scale (UCLA-LS), and the Multidimensional Scale of Perceived Social Support (MSPSS). A network analysis approach was employed to construct a network representing adolescents basic psychological needs and psychological behavioral problems, focusing on centrality measures and demographic subgroup differences. Results: A total of 418 students (20.9%) reported abnormal emotional and behavioral problems. Perceived autonomy and competence were negatively correlated with emotional problems (weights: 0.12, 0.14) and hyperactivity (weights: 0.10, 0.16). Social support showed negative correlation with peer relationship issues, hyperactivity, and conduct problems (weights: 0.16, 0.13, 0.10). Loneliness was positively correlated with emotional symptoms and peer relationship problems (weights: 0.28, 0.18). In the overall network, perceived relationships (social support and loneliness), emotional symptoms, and hyperactivity emerged as central nodes. Significant differences in network structure were observed between gender subgroups ( P =0.02). Girls internalizing issues were more influenced by loneliness and perceived autonomy frustration, while social support exhibited higher centrality in boys. Conclusions Perceived relationships, emotional problems, and hyperactivity are key nodes in the network of adolescents basic psychological needs and psychological behavioral problems. Loneliness demonstrates a prominent influence within the network, and the overall network exhibits gender differences.

Objective To explore the independent risk factors for complicated coronary heart disease (CHD) in patients with obstructive sleep apnea-hypopnea syndrome (OSAHS) and to construct a risk prediction model. Methods A retrospective study was performed to analyze the clinical data of 308 OSAHS patients who visited Jiangsu Provincial People's Hospital from January 2021 to June 2025. The patients were divided into the CHD group (n=80) and non-CHD group (n=228) according to the presence or absence of complicated CHD. Independent risk factors were identified using univariate analysis and multivariate logistic regression, and a nomogram prediction model was constructed. The model was evaluated using the calibration curve, Hosmer-Lemeshow test, Decision Curve Analysis (DCA) curve, and Receiver Operating Characteristic (ROC) curve. Results The incidence of CHD in OSAHS patients was 25.97%. Age, history of hypertension, oxygen desaturation index (ODI), severe polysomnography (PSG) grade, decreased high-density lipoprotein cholesterol, elevated homocysteine, and elevated troponin I were independent risk factors for OSAHS complicated with CHD (P<0.05). The nomogram model demonstrated good calibration (Hosmer-Lemeshow test, P=0.995), and the DCA curve showed high clinical net benefit. The area under the ROC curve was 0.940 (95% CI: 0.908~0.972), with a sensitivity of 0.842 and a specificity of 0.900. Conclusion Age, history of hypertension, ODI, PSG grade, high-density lipoprotein cholesterol, homocysteine, and troponin I are independent risk factors for CHD in OSAHS patients. The prediction model constructed based on these factors can provide a reliable tool for clinical nurses to early identify high-risk individuals and formulate targeted intervention strategies.

Objective To evaluate the degree of invasion of placental implantation by quantitative delin-eation of MRI placental bed abnormal hyperplasia vessels by 3D-Slicer software.Methods Fifty-seven pa-tients with placenta implantation requiring MRI in Quanzhou Maternal and Child Health Hospital from July 2020 to December 2023 were retrospectively collected and divided into the penetrating placenta implantation group(n=11)and the non-penetrating placenta implantation group(n=46),and the differences of MRI mor-phological signs between the two groups were analyzed.Among them,37 patients with placental bed xenogenic vessels scanned by MRI were screened and there were 11 cases in the penetrating placental implantation group and 26 cases in the non-penetrating placental implantation group.3D-Slicer software was used to delineate the surface area and volume of placental bed xenogenic vessels,the differences between the two groups were ana-lyzed.Results There were significant differences in the surface area and volume of xenogenic blood vessels in placental bed between the two groups(P＜0.05).The large surface area of xenogenic vessels in the placental bed was an independent risk factor for penetrative placental implantation,and the predicted area under the curve(AUC)of penetrative placental implantation was 0.927.When the surface area of the placental bed was 1 079.25 mm2,the sensitivity was 81.8%and the specificity was 88.5%.Conclusion According to 3D-Slicer soft-ware,the diagnostic efficacy of placental bed allogenic blood vessel surface area for penetrating placental implanta-tion is high.The larger the allogenic blood vessel surface area is,the deeper the placental invasion of uterus is.

Objective To investigate the differences in acute mountain sickness between individuals en-tering the plateau for the first time and those with multiple plateau experiences,and to examine the impact on psychological stress.Methods A total of 2 337 young plateau-acclimatized individuals from an acclimatization rotation area during November 2023 to January 2024 were selected as the study subjects.Those entering the plateau for the first time were classified into Group A(n=425),while those with previous plateau experience(at least one prior visit)were classified into Group B(n=1 912).Differences were analyzed between the two groups in terms of general characteristics,as well as acute mountain sickness(AMS)scores,Self-Rating De-pression Scale(SDS)scores,Symptom Checklist-90(SCL-90)scores,and various vital signs at the following time points:before entering the plateau(T0),on the 1st day(T1),4th day(T2),7th day(T3),and 10th day(T4)after entering the plateau.Results Group A was younger than Group B,with a statistically significant difference(P<0.05).At T1,T2,and T3,AMS scores in Group A were significantly higher than those in Group B(P<0.05).At T4,there was no statistically significant difference in AMS scores between the two groups(P>0.05).At T1,T2,and T3,blood oxygen saturation was lower and heart rate higher in Group A than in Group B,with statistically significant differences(P<0.05).After high-altitude acclimatization,no statistically significant differences in blood oxygen saturation or heart rate were observed between the groups at T4(P>0.05).At T1,there were no statistically significant differences in systolic or diastolic blood pres-sure between the groups(P>0.05).At T2,both systolic and diastolic blood pressure were higher in Group A than in Group B(P<0.05).At T3 and T4,after high-altitude acclimatization,systolic blood pressure de-creased in both groups but remained higher in Group A than in Group B(P<0.05).On the SDS scale,men-tal-emotional and somatic symptom scores were significantly higher in Group A than in Group B(P<0.001).On the SCL-90,all nine symptom factor scores were higher in Group A than in Group B(P<0.05).Conclusion Indi-viduals entering high-altitude areas for the first time experience more severe acute mountain sickness and more pronounced psychological stress.Therefore,enhanced monitoring,psychological counseling,and tailored pre-vention and treatment strategies should be implemented.

Objective To analyze the characteristics of influenza incidence and the changes of innate immune cells among children who visited the pediatric outpatient or emergency department of a hospital in Shanghai from Jan.2019 to Dec.2023,so as to provide references for the prevention of influenza in children.Methods Influenza-like illness(ILI)children who first visited the pediatric outpatient or emergency department of The First Affiliated Hospital of Naval Medical University from Jan.2019 to Dec.2023,and underwent influenza virus antigen or nucleic acid testing were enrolled.Their clinical and laboratory data were collected for retrospective analysis.With a month as the basic monitoring unit,a line chart was used to describe the changes in the number of influenza cases and the positive rate.Then,these cases were assigned to pre-coronavirus disease 2019(COVID-19)group(Jan.2019 to Dec.2019),mid-COVID-19 group(Jan.2020 to Dec.2022),or post-COVID-19 group(Jan.2023 to Dec.2023),and the influenza positive rates among these groups were compared.To further observe the changes of the innate immune cells of children with influenza A after non-pharmacological intervention(NPI)measures,considering the development characteristics of children's white blood cells,children with influenza A were assigned to 2 subgroups according to age:0-6 years or 7-16 years.Within each subgroup,they were assigned to pre-NPI group or post-NPI group according to time.A ratio of 1∶1 matching was carried out according to gender and age using propensity score matching,and the blood routine parameters were compared between the 2 groups.Results A total of 41 028 ILI children were enrolled.During the COVID-19 period from Jan.2020 to Dec.2022,the influenza positive rates decreased significantly(P＜0.001),and almost no influenza virus was detected.The peak of influenza reappeared in Mar.2023,with a lag in the peak period,but higher than previous years'peak,and the speed of reaching the peak was faster.In each subgroup,the white blood cell count and lymphocyte ratio of children with influenza A in the post-NPI group were lower than those in the pre-NPI group(all P＜0.001),and the monocyte ratio was higher than that in the pre-NPI group(both P＜0.001).Conclusion The implementation of NPI measures during COVID-19 leads to disruption of influenza transmission route and changes in the epidemic trend of influenza.The influenza virus causes serious public health problems after COVID-19 period,and children with influenza A experience more serious immune response.It is recommended to strengthen the vaccination of influenza vaccine before the flu season.

Objective To evaluate the value of a nomogram model based on ultrasonographic features and inflammatory indicators in the preoperative noninvasive prediction of pathological grading of urothelial carcinoma of bladder(UCB).Methods A retrospective analysis was conducted on 471 patients with pathologically confirmed UCB,and the patients were assigned to high-grade group(401 cases)or low-grade group(70 cases).Basic clinical data(gender,age,macroscopic hematuria),ultrasonographic features(lesion location,blood flow signal,etc.),and blood inflammatory indicators(e.g.neutrophil-to-lymphocyte ratio[NLR])were collected.Independent predictors were screened using univariate and multivariate logistic regression,and a nomogram model was constructed.Model performance was evaluated using the receiver operating characteristic(ROC)curve,calibration curve,and decision curve analysis(DCA).Results Multivariate logistic analysis identified gender(odds ratio[OR]=2.68),age(OR=1.08),macroscopic hematuria(OR=3.19),lesion located in the trigone(OR=4.59),positive blood flow signal(OR=2.87),and NLR(OR=1.03)were independent predictors of high-grade UCB(all P＜0.05).The combined model(clinical features+ultrasonographic characteristics+inflammatory indicators)achieved an area under curve(AUC)of 0.892,which was significantly higher than the clinical feature-only model(AUC=0.799)and the clinical+ultrasonographic model(AUC=0.856).The calibration curve demonstrated good consistency between predicted and actual outcomes,and DCA confirmed its optimal clinical net benefit.Conclusion The nomogram model integrating clinical features,ultrasonographic characteristics,and inflammatory indicators can effectively discriminate UCB pathological grading,providing a reliable preoperative noninvasive assessment tool for personalized treatment decisions.

Objective To discuss the application of ankle pump exercise with colorful ring counting command in improving the blood hypercoagulable state in patients after hepatic arterial infusion chemotherapy(HAIC).Methods A total of 105 patients,who received HAIC at the Interventional Department of a certain Grade A general hospital in Fujian Province of China from January to December of 2023,were collected.The 53 patients who received treatment from January to June of 2023 were used as the control group,and the other 52 patients who received treatment from July to December of 2023 were used as the study group.The postoperative routine nursing and ankle pump exercise was employed for the patients of the control group,while ankle pump exercise with colorful ring counting command on the basis of postoperative routine nursing was carried out for the patients of the study group.The resting status of popliteal vein,the maximum and average systolic blood flow velocity at the end of exercise,the changes in coagulation index,the incidence of deep vein thrombosis,and the compliance of ankle pump were compared between the two groups.Results There were no statistically significant differences in the resting popliteal vein blood flow velocity and incidence of deep vein thrombosis between the two groups(P＞0.05).The maximum and average popliteal vein blood flow velocity at the end of exercise in the study group were higher than those in the control group.The levels of PT,APTT,FBG,TT and D-dimer(D-D)at 72 hours after HAIC were lower than those in the control group(all P＜0.05),and the compliance of ankle pump exercise in the study group was better than that in the control group(P＜0.05).Conclusion Ankle pump exercise with colorful ring counting command can accelerate lower limb blood flow velocity,improve the blood hypercoagulable state,and improve compliance with ankle pump exercise in patients after receiving HAIC.