目的 探讨血清乳酸脱氢酶（LDH）在中晚期肝癌患者接受靶向联合免疫治疗后的预后预测价值。方法 选取2022年1月—2024年8月在莆田学院附属医院肿瘤内科经病理和影像学检查确诊的中晚期肝癌患者作为研究对象。从医院的电子病历系统中收集患者的基线资料，随访截止2025年8月，并记录随访结果，包括患者的疾病缓解情况和死亡情况，以及无疾病进展生存期（PFS）、总生存期（OS）。采用Kaplan-Meier方法绘制不同基线LDH水平患者的OS生存曲线，并通过Log-rank检验比较生存曲线。同时，运用多因素Cox比例风险回归分析探讨影响中晚期肝癌患者在接受靶向联合免疫治疗后OS的相关因素。结果 结果显示，在50例肝癌患者中，基线LDH低于200 U/L的有15例，而高于200 U/L的有35例。与基线LDH＜200 U/L组相比，基线 LDH≥200 U/L患者PFS、OS更短，差异均有统计学意义（χ²分别为5.51、15.6，P值分别为0.019、0.017）。治疗8周后，与LDH降低患者相比，LDH升高患者OS更短，差异有统计学意义（χ²=13.2，P=0.04）。多因素Cox比例风险回归分析结果表明，基线LDH水平超过200 U/L是中晚期肝癌患者接受靶向联合免疫治疗后OS的影响因素［P=0.035，HR（95%CI）=5.03（1.12，22.54）］。结论 基线LDH水平较低的患者表现出更好的OS。基线LDH水平可以作为预测中晚期肝癌患者在接受靶向联合免疫治疗时预后的指标。 

   Objective To evaluate the prognostic significance of serum lactate dehydrogenase（LDH）levels in patients with advanced hepatocellular carcinoma（HCC）undergoing targeted therapy combined immunotherapy．Methods Patients diagnosed with advanced HCC were selected in Putian College Affiliated Hospital from January 2022 to August 2024，diagnosed with pathological and imaging examinations results．Patient baseline data were collected from the hospital’s electronic medical records，with follow-up extending until August 2025．We documented outcomes such as disease response and mortality，along with progression-free survival（PFS）and overall survival（OS）．Kaplan-Meier survival curves were constructed based on baseline LDH levels，and the Log-rank test was employed for comparison．Additionally，multivariate Cox proportional hazards regression analysis was conducted to identify factors influencing OS in patients receiving targeted therapy combined immunotherapy．Results Among the 50 patients，15 had baseline LDH levels below 200 U/L，while 35 had levels above．Patients with baseline LDH≥200 U/L had significantly shorter PFS and OS than those with baseline LDH ＜200 U/L（χ²=5.51 and 15.6 for PFS and OS，respectively；P=0.019 and 0.017，respectively）．After 8 weeks of treatment，patients with increased LDH had significantly shorter OS compared with patients with decreased LDH（χ²=13.2，P=0.04）．Multivariate Cox proportional hazards regression analysis indicated that a baseline LDH level exceeding 200 U/L is an independent prognostic factor for OS in patients with intermediate to advanced HCC receiving targeted therapy combined with immunotherapy（P=0.035，HR 5.03［1.12，22.54］）．Conclusions Patients with lower baseline LDH levels demonstrated better OS，suggesting that baseline LDH can serve as an important prognostic indicator for advanced HCC patients undergoing targeted combined immunotherapy．

       目的 构建首发脑出血患者并发卒中相关性肺炎的风险预测模型并验证模型的预测性能。方法 回顾性分析2012年1月—2022年12月广州市第一人民医院治的419例首发脑出血患者的临床资料，按照7︰3比例随机化分为训练列（293例）和验证队列（126例）。统计基于开发队列数据，采用Logistic回归模型分析首发脑出血患者并发卒中相关性肺炎的影响因素，并构建风险预测模型。基于开发队列和验证队列数据，采用校准曲线、受试者操作特征（ROC）曲线下面积和决策曲线分析模型的预测性能。结果 419例首发脑出血患者中有113例发生卒中相关性肺炎，发生率为26.97%。美国国立卫生研究院卒中量表（NIHSS）评分、吞咽困难、初始血肿体积、中性粒细胞百分比与白蛋白比值（NPAR）、中性粒细胞计数与淋巴细胞计数比值（NLR）、手术治疗、气管插管、留置胃管均是首发脑出血患者并发卒中相关性肺炎的影响因素（P＜0.05）。基于上述影响因素构建了首发脑出血患者并发卒中相关性肺炎的风险预警模型，校准曲线显示模型在开发队列和验证队列中预测卒中相关性肺炎发生率均与实际发生率相近；ROC曲线显示此模型在开发队列、验证队列中预测的曲线下面积分别为0.906（95%CI：0.867~0.937）、0.884（95%CI：0.815~0.934）；决策曲线分析显示当开发队列阈概率在3%~80%内、验证队列阈概率在2%~76%内使用此模型干预比全/无干预更有临床价值。结论 基于NIHSS评分、吞咽困难、初始血肿体积、NPAR、NLR、手术治疗、气管插管、留置胃管构建的首发脑出血患者并发卒中相关性肺炎的风险预测模型具有良好预测性能和临床应用价值。

      Objective To construct a risk prediction model for stroke associated pneumonia in patients with initial cerebral hemorrhage（ICH）and validate the predictive performance of the model．Methods A retrospective analysis was conducted on the clinical data of 419 patients with ICH admitted to our hospital from January 2012 to December 2022．They were randomly divided into a development cohort（293 cases）and a validation cohort（126 cases）according to a 7∶3 ratio．The Logistic regression model was used to analyze the influencing factors of stroke related pneumonia in patients with ICH based on the development cohort data，and a risk prediction model was constructed．Based on the development cohort data and validation cohort data，the predictive performance of the model was analyzed using calibration curves，receiver operating characteristic（ROC）curve，and decision curve analysis．Results Among 419 patients，113 developed stroke associated pneumonia，with a rate of 26.97%．The National Institutes of Health Stroke Scale（NIHSS）score，swallowing difficulties，initial hematoma volume，neutrophil percentage to albumin ratio（NPAR），neutrophil count to lymphocyte count ratio（NLR），surgical treatment，endotracheal intubation，and indwelling gastric tube were all independent influencing factors for stroke associated pneumonia in patients with ICH（P＜0.05）．Based on the above influencing factors，a risk prediction model for stroke associated pneumonia in patients with ICH was constructed．The calibration curve showed that the predicted incidence of stroke associated pneumonia by the model in both the development and validation cohorts was close to the actual incidence．The ROC curve showed that the predicted area under the curve for this model in the development cohort and validation cohort was 0.906（95%CI：0.867-0.937）and 0.884（95%CI：0.815-0.934），respectively．The decision curve analysis showed that when the threshold probability of the development cohort was between 3%-80%，and the threshold probability of the validation cohort was between 2%-76%，the intervention using this model was more clinically valuable than all/no intervention．

Conclusions The risk prediction model for stroke associated pneumonia in patients with ICH based on NIHSS score，swallowing difficulties，initial hematoma volume，NPAR，NLR，surgical treatment，tracheal intubation，and indwelling gastric tube has good predictive performance and clinical application value．

       目的 构建与验证回肠造口术后造口周围皮炎发生的列线图预测模型。方法 回顾性分析广州市中山大学附属第六医院2022年7月—2023年7月收治的441例回肠造口术患者资料。通过单因素Logistic回归确定影响因素，基于训练队列建立列线图模型，并采用受试者操作特征（ROC）曲线和校准曲线分别评估模型的区分度与校准能力，同时通过决策曲线分析（DCA）评估临床实用性。在内部验证队列进行模型性能验证。结果 单因素Logistic回归确定6个危险因素，最终选取其中5个（是否糖尿病、造口定位、是否接受造口护理宣教、造口皮肤凹陷、性别）构建预测模型。训练队列中，模型校准曲线显示良好拟合优度（Hosmer and Lemeshow检验P = 0.976），ROC曲线下面积为0.672；决策曲线证实模型具有临床应用价值，预测模型在验证集上有类似结果。结论 本研究建立的5参数列线图模型具有良好的区分度、校准能力和临床实用价值，可为造口周围皮炎的预测提供量化工具。

       Objective To establish a predictive nomogram for peristomal dermatitis after ileostomy．Methods This study analyzed 441 patients who underwent ileostomy at a third-class premium hospital．The risk factors of peristomal dermatitis were determined by univariate binary Logistics regression．Then a predictive nomogram was established with the risk factors in the training cohort．The discrimination and calibration ability of the nomogram was demonstrated by receiver operating characteristic curves and calibration curves，respectively．Besides，a testing cohort was utilized to validate the nomogram model．To reflect the clinical utility of the model，we also performed decision curve analysis．Results A total of six parameters was determined as the risk factors of peristomal dermatitis．Finally，five of them（whether diabetes，stoma location，whether to receive stoma care education，stoma skin depression，gender）were chosen to establish the nomogram．The calibration curve showed favorable calibration ability in the training cohort with a Hosmer and Lemeshow goodness of fit test P-value of 0.976 and the area under the curve of the nomogram is 0.672．The decision curve showed nomogram had potential clinical utility．These results were consistent with the testing cohort．Conclusions The nomogram established by five parameters was capable of predicting the occurrence of peristomal dermatitis with favorable discrimination and calibration ability and shows potential clinical utility．

       目的 评估使用炎症、免疫及营养指标对于胃癌根治术后发生腹膜转移的预测作用，并建立适用于术前风险评估的预测模型。方法 回顾性收集2022—2024年苏州市立医院85例晚期胃癌患者的临床资料，按术后是否发生腹膜转移分为转移组44例与非转移组41例。通过分析比较两组间的营养参数［血清白蛋白、纤维蛋白原及预后营养指数（PNI）］、炎症指标［中性粒细胞计数、C反应蛋白、中性粒细胞与淋巴细胞比值（NLR）及全身免疫炎症指数（SII）］、肿瘤标志物以及免疫功能指标（淋巴细胞计数、IL-2、IL-6、TNF-α和IFN-γ）等。使用单因素及多因素Logistic回归分析确定独立预测因子，并构建预测模型，采用Bootstrap法内部验证，计算校正后的曲线下面积及收缩因子，以优化模型性能。结果 多因素分析结果显示，IL-6（OR=1.116，95%CI：1.020~1.224）与CA125（OR=1.016，95%CI：1.006~1.026）为术后腹膜转移的独立预测因素，构建的预测模型区分能力良好，初始AUC为0.801（95%CI：0.702~0.900），经Bootstrap校正后AUC为0.797，模型校准度理想，具有良好的临床稳健性。结论 基于术前IL-6和CA125水平的预测模型，可用于评估胃癌患者根治术后发生腹膜转移的风险。该模型可在术前阶段有效识别高风险病例，为临床制定个体化辅助治疗方案及随访策略提供关键参考，并具有改善患者预后的潜在价值。

       Objective To evaluate the predictive value of inflammatory，immune，and nutritional indicators for peritoneal metastasis after radical gastrectomy in gastric cancer and to establish a predictive model applicable for preoperative risk assessment．Methods Clinical data from 85 patients with advanced gastric cancer treated at Suzhou Municipal Hospital between 2022 and 2024 were retrospectively collected．Patients were divided into two groups based on postoperative peritoneal metastasis occurrence：a metastasis group（n=44）and a non-metastasis group（n=41）．Indicators analyzed and compared between groups included nutritional parameters（serum albumin，fibrinogen，PNI），inflammatory markers（neutrophil count，CRP，NLR，SII），tumor markers and immune indicators（lymphocyte count，IL-2，IL-6，TNF-α，IFN-γ）．Univariate and multivariate logistic regression analyses identified independent predictors for constructing the prediction model．Internal validation was performed using 1 000 bootstrap resamples to calculate the adjusted area under the curve（AUC）and shrinkage factor for model optimization．Results Multivariate analysis identified IL-6（OR=1.116，95%CI：1.020-1.224）and CA125（OR=1.016，95%CI：1.006-1.026） as independent predictors of postoperative peritoneal metastasis．The constructed predictive model demonstrated good discriminatory ability，with an initial AUC of 0.801（95%CI：0.702–0.900）and a bootstrap-corrected AUC of 0.797．The model exhibited ideal calibration and good clinical robustness．Conclusions This study established a predictive model based on preoperative IL-6 and CA125 levels to assess the risk of peritoneal metastasis after radical surgery in gastric cancer patients．The model effectively identifies high-risk cases preoperatively，providing critical guidance for developing individualized adjuvant treatment and follow-up strategies，with potential to improve patient prognosis．

       目的 探讨自回归求和时间序列模型（ARIMA）在耐碳青霉烯类肺炎克雷伯菌（CRKP）医院感染预测中的应用，为CRKP感染的科学防控提供策略。方法 选取贾汪区人民医院住院患者2018年1月—2024年12月共7年每月CRKP医院感染率，应用SPSS 19.0建立ARIMA模型，利用该模型对2024年1—12月每月CRKP医院感染数据进行验证，以评价模型的预测性能。结果 用2018—2024年住院患者每月CRKP医院感染率建模、拟合，建立最优模型ARIMA（1，1，1），模型拟合值与实际值较吻合，模型对CRKP医院感染率实际值与预测值吻合度较高，平均相对误差为5.40%。结论 用ARIMA模型可有效拟合、预测CRKP医院感染情况，为CRKP防控提供科学指导。

     Objective To explore the value of auto regressive integrated moving average（ARIMA）model in predicting theinfection rate of carbapenem-resistant Klebsiella pneumoniae（CRKP）．Methods The monthly incidence of nosocomial infection of CRKP in Jiawang District People’s Hospital from 2018 to 2024 was selected，and the ARIMA model was established by19.0 SPSS to analyze the fitting and prediction value of the model．Results The ARIMA was established based on the monthly CRKP infection rate from January 2018 to December 2024．The fitted value of the ARIMA（1，1，1）model was in good agreement with the actual value．The incidence of CRKP infection were in good agreement with the predicted value．The average relative errors were 5.40%．Conclusions The ARIMA model can effectively fit and predict the CRKP infection rate，providing scientific guidance for the prevention and control of CRKP infection．

      目的 探讨多模态深度学习融合心脏超声与心电图特征对冠心病患者心源性猝死的预测价值。方法 选取2024年1月—2025年6月收治的60例冠心病患者，所有患者均接受心脏超声、心电图检查及多模态深度学习模型预测，随访6个月记录心源性猝死事件。根据随访结果分为事件组（n=15）和非事件组（n=45），比较两组临床资料及各项指标差异。结果 多模态深度学习模型预测敏感度为86.67%，特异度为91.11%，准确率为90.00%，AUC为0.923，显著优于单独心脏超声（AUC=0.761）和单独心电图（AUC=0.788）。结论 多模态深度学习融合心脏超声与心电图特征能够有效预测冠心病患者心源性猝死风险，预测性能优于传统单一模态评估方法。

      Objective To investigate the predictive value of multimodal deep learning integrating echocardiography and electrocardiogram features for sudden cardiac death in patients with coronary heart disease．Methods A total of 60 patients with coronary heart disease admitted from January 2024 to June 2025 were enrolled．All patients underwent echocardiography，electrocardiogram examination，and multimodal deep learning model prediction，with a 6-month follow-up to record sudden cardiac death events．According to the follow-up results，patients were divided into the event group（n=15） and non-event group（n=45），and clinical data and various indicators were compared between the two groups．Results The multimodal deep learning model achieved a sensitivity of 86.67%，specificity of 91.11%，accuracy of 90.00%，and AUC of 0.923，which were significantly superior to echocardiography alone（AUC=0.761） and electrocardiogram alone（AUC=0.788）．Conclusions Multimodal deep learning integrating echocardiography and electrocardiogram features can effectively predict the risk of sudden cardiac death in patients with coronary heart disease，with predictive performance superior to traditional single-modality assessment methods．

       目的 探究1990—2023年中国人群下背痛（LBP）疾病负担，并预测2024—2040年LBP疾病负担发展趋势，为公共卫生政策制定提供数据支撑。方法 基于GBD2023数据库，采用Joinpoint模型、分解分析（decomposition analysis）量化我国LBP负担变化趋势，最后通过ARIMA（autoregressive integrated moving average model）模型预测未来趋势。结果 2023年我国LBP患病人数达9 532.40万人，发病人数为4 138.36万人，伤残调整生命年（DALYs），虽然患病人数、发病人数及DALYs绝对数较1990年显著增长，但年龄标准化率（ASR）均呈显著下降趋势（AAPC依次为-0.90%、-0.84%、-0.89%）。分解分析表明，人口年龄结构是我国LBP疾病负担增长的主要正向驱动因素，流行病学趋势变化对患病率和DALYs起关键抑制作用，且存在明显性别差异。ARIMA模型预测显示，2024—2040年男性和女性LBP相关ASR将持续下降，但我国女性LBP负担仍高于男性。结论 我国LBP负担沉重，受人口结构、性别差异、行为与职业等多方面因素影响，未来需通过分人群、分职业精准干预及多部门协同，持续降低疾病负担。

     Objective To explore the characteristics of disease burden of low back pain（LBP） in the Chinese population from 1990 to 2023，and predict its development trend from 2024 to 2040，so as to provide data support for the formulation of public health policies．Methods Based on the Global Burden of Disease Study 2023（GBD 2023） database，the Joinpoint model and decomposition analysis were adopted to quantify the changing trend of LBP burden in China．The autoregressive integrated moving average（ARIMA） model was used to predict the future trend，with the data for prediction stratified by gender and age group．Results In 2023，the number of LBP patients in China reached 95.3240 million，the number of incident cases was 41.3836 million，and the disability-adjusted life years（DALYs） were 10.6359 million person-years．Although the absolute numbers of prevalent cases，incident cases，and DALYs increased significantly compared with 1990，the age-standardized rates（ASRs） all showed a significant downward trend（average annual percentage changes were -0.90%，-0.84%，and -0.89%，respectively）．Decomposition analysis revealed that population age structure was the primary positive driving factor for the growth of LBP disease burden in China，while changes in epidemiological trends played a key inhibitory role in prevalence and DALYs，with significant gender differences．ARIMA model prediction indicated that the LBP-related ASRs would continue to decrease in both males and females from 2024 to 2040，but the LBP burden in females would remain higher than that in males in China．Conclusions The baseline burden of LBP in China is heavy，affected by multiple factors such as population structure，gender differences，behavioral and occupational factors．In the future，targeted interventions for specific populations and occupations，as well as multi-sectoral collaboration，are needed to continuously reduce the disease burden．

      目的 探讨预后营养指数（PNI）对脓毒症合并心力衰竭患者发生谵妄的预测价值，并评估其在指导护理风险分层中的作用。方法 采用回顾性队列研究，纳入2022年2月—2024年9月广东省第二中医院收治的162例脓毒症合并心力衰竭患者。通过ROC曲线确定PNI预测谵妄的最佳截断值，并以此将患者分为高低PNI组。比较两组基线特征，采用Kaplan-Meier曲线分析谵妄累积发生率和28天院内死亡率，通过Cox比例风险模型评估PNI与谵妄及预后的独立关联，并利用ROC曲线评价其预测性能。最后利用亚组分析技术进一步评估PNI预测价值的普适性。结果 低PNI（≤35.72）是谵妄发生的独立危险因素。竞争风险模型显示，低PNI组10 d谵妄累积发生率达37.8%，高于高PNI组的10.0%。多因素分析表明，PNI每降低一个单位，谵妄风险增加约17.2%。限制性立方样条分析提示PNI与谵妄风险呈线性负相关。ROC曲线分析显示PNI预测谵妄的AUC为0.721，其预测效能优于白蛋白和SOFA评分。亚组分析进一步证实PNI的保护效应在不同临床特征患者中具有一致性。结论 PNI是脓毒症合并心力衰竭患者谵妄风险的独立预测指标，其预测效能优于传统单一指标。基于PNI的风险分层有助于早期识别谵妄高危患者，为实现精准护理干预提供重要依据。

      Objective To explore the predictive value of the prognostic nutritional index（PNI）for delirium in patients with sepsis complicated by heart failure，and to evaluate its role in guiding nursing risk stratification．Methods A retrospective cohort study was conducted，including 162 patients with sepsis and heart failure admitted to Guangdong Provincial Second Hospital of Traditional Chinese Medicine between February 2022 and September 2024．The optimal PNI cut-off value for predicting delirium was determined using ROC curve analysis，and patients were divided into high and low PNI groups accordingly．Baseline characteristics of the two groups were compared．Kaplan-Meier curves were used to analyze the cumulative incidence of delirium and 28-day in-hospital mortality．Cox proportional hazards models were employed to assess the independent association between PNI and delirium as well as prognosis，and ROC curves were used to evaluate its predictive performance．Subgroup analysis was further utilized to assess the generalizability of PNI’s predictive value．Results A low PNI（≤35.72）is an independent risk factor for delirium．The competing risk model showed that the cumulative incidence of delirium in the low PNI group was 37.8% at 10 days，significantly higher than the 10.0% in the high PNI group．Multivariate analysis indicated that for each unit decrease in PNI，the risk of delirium increased by 17.2% approximately．Restricted cubic spline analysis suggested a linear negative correlation between PNI and delirium risk．ROC curve analysis demonstrated that the AUC of PNI for predicting delirium was 0.721，with its predictive performance surpassing that of albumin and SOFA score．Subgroup analysis further confirmed that the protective effect of PNI was consistent across patients with different clinical characteristics．Conclusions PNI is an independent predictor of delirium risk in patients with sepsis and heart failure，with superior predictive performance compared to traditional single indicators．Risk stratification based on PNI aids in the early identification of patients at high risk for delirium，providing an important basis for precise nursing interventions．

       肺癌作为全球恶性肿瘤中发病率和死亡率较高的一种，手术是其主要治疗手段。然而，肺切除手术常引起肺功能下降，影响患者术后功能恢复及生活质量。准确预测术后肺功能对制定个体化手术方案、降低并发症风险及改善患者预后具有重要意义。目前，研究者已开发出多种结合临床指标、影像学参数及生物标志物的预测模型，用于评估肺癌患者术后肺功能变化。然而，现有模型在准确性、特异性、标准化及临床推广方面仍存在诸多挑战。本文系统综述肺癌术后肺功能的影响因素，并对现有预测模型进行总结与评价，以期为患者术前评估、手术方案选择及术后管理提供参考，进而为改善治疗效果与患者生活质量提供理论依据。

       Lung cancer is a kind of malignant tumor with high morbidity and mortality in the world，and surgery is the main treatment．However，pneumonectomy is often accompanied by the loss of lung function，which seriously affects the recovery of lung function and the  quality of life．Accurate prediction of postoperative pulmonary function is of great significance for formulating individualized surgical plans，reducing the risk of complications and improving the prognosis of patients．At present，researchers have developed a variety of predictive models that combine clinical indicators，imaging parameters，and biomarkers to evaluate postoperative lung function changes in patients with lung cancer．However，there are still many challenges in the accuracy，specificity，standardization and clinical promotion of existing models．This paper systematically reviews the influencing factors of lung function after lung cancer surgery，and summarizes and evaluates the existing prediction models，in order to provide a reference for preoperative evaluation，surgical plan selection and postoperative management of patients，and then provide theoretical basis for improving treatment effect and quality of life of patients．

       目的  探讨结肠镜下息肉切除术后复发的危险因素，并基于机器学习算法构建复发风险预警模型，为防治对策提供依据。方法  回顾性收集2018年9月—2023年9月六安市人民医院1 058例初次行无痛结肠镜下息肉切除术患者的临床资料，使用单因素和多因素Logistic回归分析筛选复发危险因素。采用7∶3随机抽样分为训练集和验证集，分别通过决策树、贝叶斯及Logistic回归算法构建预测模型，并以受试者工作特征曲线（ROC）曲线下面积（AUC）、灵敏度、特异度等指标来评估模型效能。结果  单因素分析显示，性别、吸烟、代谢综合征、息肉数量、息肉位置、山田分型、组织病理学类型、切除方式、复查时间、肠息肉直径、手术时间是复发的危险因素（P＜0.05）。多因素分析显示，性别、代谢综合征、息肉数量、息肉直径、肠息肉位置、山田分型、组织学病理类型、切除方式、手术时间均是结肠息肉内镜下切除术后复发的危险因素。模型评估显示，决策树算法、贝叶斯算法、Logistic回归算法的ROC曲线下面积（AUC）分别为0.849、0.818、0.811；灵敏度分别为85.14%、81.62%、79.43%；特异度分别为81.69%、79.45%、74.18%；约登指数分别为0.534、0.551、0.573；95%CI分别为0.810~0.876、0.794~0.860、0.782~0.850；决策树算法模型效能最佳，Logistic回归算法的性能最差。结论  性别、代谢综合征、肠息肉特征（数量、直径、位置等）是术后复发的关键危险因素。决策树模型在风险预测中表现最优，可为临床制定个体化随访策略提供参考。

       Objective  To explore the  risk factors for  recurrence after painless colonoscopic polypectomy and construct a recurrence risk warning model based on machine learning algorithms to provide evidence for prevention and treatment strategies．Methods  A retrospective analysis was conducted on clinical data from 1 058 patients who underwent their first painless colonoscopy-guided polypectomy at our hospital between September 2018 and September 2023．Univariate and multivariate Logistic  regression analyses were performed to identify recurrence risk factors．The dataset was randomly divided into training and validation sets using a 7∶3 ratio．Prediction models were constructed using decision tree，Bayesian，and Logistic regression algorithms，and their performance was evaluated using metrics such as the area under the receiver operating characteristic curve（AUC），sensitivity，specificity，and others．Results  Univariate analysis revealed that gender，smoking，metabolic syndrome，number of polyps，polyp location，Yamada classification，histopathological type，resection method，follow-up time，polyp diameter，and operation duration were risk factors for recurrence（P＜0.05）．Multivariate analysis identified gender，metabolic syndrome，number of polyps，polyp diameter，polyp location，Yamada classification，histopathological type，resection method，and operation duration as independent risk factors for recurrence after endoscopic polypectomy．Model evaluation showed AUC values of 0.849，0.818，and 0.811 for the decision tree，Bayesian，and Logistic regression algorithms，respectively．Sensitivity values were 85.14%，81.62%，and 79.43%；specificity values were 81.69%，79.45%，and 74.18%；Youden’s indices were 0.534，0.551，and 0.573；and 95% confidence intervals（CIs）were 0.810–0.876，0.794–0.860，and 0.782–0.850，respectively．The  decision tree algorithm demonstrated the best predictive performance，while the Logistic regression algorithm performed the least favorably．Conclusions  Gender，metabolic syndrome，and polyp characteristics（number，diameter，location，etc．）are key  risk factors for recurrence after polypectomy．The decision tree algorithm exhibited optimal predictive efficacy，offering valuable insights for developing individualized follow-up strategies in clinical practice．