切换至 "中华医学电子期刊资源库"
高级检索
中华腔镜泌尿外科杂志(电子版)

中华腔镜泌尿外科杂志(电子版) ›› 2026, Vol. 20 ›› Issue (03) : 297 -306. doi: 10.3877/cma.j.issn.1674-3253.2026.03.010

临床研究

基于临床和CT影像组学特征的机器学习模型对经皮肾镜术后脓毒症的预测价值
徐宏博1, 胡玉良2, 魏雪栋3, 金李晨3, 武克风1, 陈逸伦1, 陆兵1, 周守军1, 侯建全1,†()   
  1. 1215006　江苏，苏州大学附属第四医院泌尿外科
    2230031　安徽，合肥市第三人民医院泌尿外科
    3215006　江苏，苏州大学附属第一医院泌尿外科
  • 收稿日期:2025-08-29 出版日期:2026-06-01
  • 通信作者: 侯建全
  • 基金资助:
    苏州市医学创新应用研究项目(SZM2023034)

The predictive value of machine learning models based on clinical and CT radiomics features for sepsis after percutaneous nephrolithotomy

Hongbo Xu1, Yuliang Hu2, Xuedong Wei3, Lichen Jin3, Kefeng Wu1, Yilun Chen1, Bing Lu1, Shoujun Zhou1, Jianquan Hou1,†()   

  1. 1Department of Urology, the Fourth Affiliated Hospital of Soochow University,Jiangsu 215006,China
    2Department of Urology, the Third People's Hospital of Hefei, Anhui 230031, China
    3Department of Urology, the First Affiliated Hospital of Soochow University, Jiangsu 215006, China
  • Received:2025-08-29 Published:2026-06-01
  • Corresponding author: Jianquan Hou
全文 ( ) 下载PDF ( ) 导出引用
引用本文：

徐宏博, 胡玉良, 魏雪栋, 金李晨, 武克风, 陈逸伦, 陆兵, 周守军, 侯建全. 基于临床和CT影像组学特征的机器学习模型对经皮肾镜术后脓毒症的预测价值[J/OL]. 中华腔镜泌尿外科杂志(电子版), 2026, 20(03): 297-306.

Hongbo Xu, Yuliang Hu, Xuedong Wei, Lichen Jin, Kefeng Wu, Yilun Chen, Bing Lu, Shoujun Zhou, Jianquan Hou. The predictive value of machine learning models based on clinical and CT radiomics features for sepsis after percutaneous nephrolithotomy[J/OL]. Chinese Journal of Endourology(Electronic Edition), 2026, 20(03): 297-306.

目的

探讨基于临床和影像组学特征的机器学习模型对经皮肾镜（PCNL）术后脓毒症发生风险的预测价值，为临床早期预防和干预提供科学依据。

方法

回顾性分析2020年1月至2024年1月339例接受PCNL治疗的肾结石患者资料。根据术后24 h的快速序贯器官衰竭评分（qSOFA）将患者分为脓毒症组和非脓毒症组。收集患者临床资料并从CT图像中提取影像组学特征，经过多步特征筛选策略后保留8个具有显著预测价值的影像组学特征。纳入患者采用7∶3的随机分割方案，分为训练集（n=237）和测试集（n=102），使用逻辑回归（LR）、支持向量机（SVM）、K近邻算法（KNN）、随机森林（RF）、极端随机树（ExtraTrees）、极端梯度提升（XGBoost）、轻量梯度提升机（LightGBM）、多层感知机（MLP）8种机器学习算法分别构建临床模型和影像组学模型，将效能最优的两个模型进行融合，建立脓毒症临床-影像组学联合模型，并构建列线图。通过绘制校准曲线评估列线图的校准效能。绘制决策曲线分析评估预测模型的临床实用性。通过受试者工作特征（ROC）曲线下面积（AUC）评估模型性能。

结果

本研究339例患者中共有24例（7.1%）术后发生脓毒症。两组比较发现，性别、术前疑似感染性结石、结石负荷、血红蛋白、尿亚硝酸盐、尿白细胞、血总蛋白、血球蛋白和血前白蛋白等指标差异有统计学意义（P<0.05）。进一步行多因素Logistic回归显示，术前疑似感染性结石（OR=5.589，95%CI：1.659~18.834，P=0.005）、尿亚硝酸盐阳性（OR=5.312，95%CI：1.802~15.662，P=0.002）和血总蛋白<68.1 g/L（OR=0.245，95%CI：0.071~0.846，P=0.026）是PCNL术后脓毒症的独立危险因素，因此我们将上述三个指标作为构建临床模型的参数。在临床模型的测试集中，ExtraTrees模型在8种机器学习模型中的表现最佳（AUC=0.743）；在影像组学模型的测试集中，KNN模型在8种机器学习模型中的表现最佳（AUC=0.878）。将上述两个模型进行融合，建立脓毒症临床-影像组学联合模型，并构建列线图。结果显示，在测试集中，结合了临床和影像组学特征的联合模型诊断效能最佳，AUC达0.898。

结论

分别基于ExtraTrees算法和KNN算法的临床-影像组学联合模型能有效预测PCNL术后脓毒症的发生风险，有助于临床医师早期识别高危患者，及时干预，降低术后脓毒症发生率。

关键词: 肾结石, 经皮肾镜, 脓毒症, 影像组学, 机器学习
Objective

To explore the predictive value of machine learning models based on clinical and radiomics features for the risk of sepsis after percutaneous nephrolithotomy (PCNL), and to provide a scientific basis for early prevention and intervention in clinical practice.

Methods

A retrospective analysis was conducted on the data of 339 patients with kidney stones who underwent PCNL from January 2020 to January 2024. Patients were divided into sepsis and non-sepsis groups based on the quick sequential organ failure assessment (qSOFA) score at 24 h postoperatively. Clinical data were collected and radiomics features were extracted from CT images. After a multi-step feature selection strategy, 8 radiomics features with significant predictive value were retained. Patients were randomly divided into a training set (n=237) and a test set (n=102) at a ratio of 7:3. Eight machine learning algorithms, namely logistic regression (LR), support vector machine (SVM), k-nearest neighbors (KNN), random forest (RF), extremely randomized trees (ExtraTrees), extreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and multi-layer perceptron (MLP), were used to construct the clinical model and radiomics model. Subsequently, the two models with the highest performance were integrated to develop a combined clinical-radiomics model for sepsis prediction, and a nomogram was established accordingly. Calibration curve was drawn to evaluate the calibration efficacy of the nomogram, and Hosmer-Lemeshow fitting analysis was used to evaluate the calibration ability of the nomogram. Decision curve analysis (DCA) was drawn to evaluate the clinical utility of the prediction model. Model performance was assessed by the area under the receiver operating characteristic (ROC) curve (AUC).

Results

Among the 339 patients in this study, 24(7.1%) developed sepsis postoperatively. Comparisons between the two groups revealed significant differences in gender, preoperative suspicion of infectious stones, stone burden, hemoglobin, positive urine nitrite, urine white blood cells, total blood protein, globulin, and prealbumin (P<0.05). Multivariate logistic regression further showed that preoperative suspicion of infectious stones (OR=5.589, 95%CI: 1.659-18.834, P=0.005), positive urine nitrite (OR=5.312, 95%CI: 1.802-15.662, P=0.002), and total blood protein <68.1 g/L (OR=0.245, 95%CI: 0.071-0.846, P=0.026) were independent risk factors for postoperative sepsis after PCNL. Consequently, these three indicators were selected as parameters for the construction of a clinical prediction model. In the test set of the clinical model, the ExtraTrees model exhibited superior performance among the eight machine learning models, achieving an AUC of 0.743. In the test set of the radiomics model, the KNN model demonstrated the highest performance among the eight machine learning models, with an AUC of 0.878. By integrating the aforementioned clinical and radiomics model, a combined clinical-radiomics model for sepsis prediction was developed, and a corresponding nomogram was constructed. The results indicated that in the test set, the integrated model incorporating both clinical and radiomics features achieved the best diagnostic performance, with an AUC of 0.898.

Conclusion

The clinical-radiomics combined model based on the ExtraTrees algorithm and the KNN algorithm can effectively predict the risk of postoperative sepsis after PCNL, which is helpful for clinicians to identify high-risk patients early and intervene in a timely manner, thereby reducing the incidence of postoperative sepsis.

Key words: Kidney stone, Percutaneous nephrolithotomy, Sepsis, Radiomics, Machine learning
表1 脓毒症与非脓毒症患者临床指标比较
项目 脓毒症组(n=24) 非脓毒症组(n=315) 统计值 P值
年龄[岁，M(Q)] 53(43，59) 52(41，58) Z=-0.168 0.867
性别[例(%)]     χ2=13.336 <0.001
男性 7(29.2) 206(65.4)    
女性 17(70.8) 109(34.6)    
体质量指数[kg/m2，M(Q)] 25.3(20.6，28.1) 24.6(22.4，26.7) Z=-0.223 0.824
结石负荷[mm2，M(Q)] 507(335，1052) 359(225，551) Z=-2.877 0.004
术前疑似感染性[例(%)] 8(33.3) 22(7.0) χ2=14.899 <0.001
多发结石[例(%)] 15(62.5) 166(52.7) χ2=0.861 0.353
肾积水[例(%)] 14(58.3) 164(52.1) χ2=0.352 0.553
高血压[例(%)] 6(25.0) 118(37.5) χ2=1.493 0.222
糖尿病[例(%)] 4(16.7) 35(11.1) χ2=0.676 0.501
既往泌尿系手术史[例(%)] 11(45.8) 120(38.1) χ2=0.563 0.453
尿细菌培养阳性[例(%)] 8(33.3) 65(20.6) χ2=2.128 0.145
尿亚硝酸盐阳性[例(%)] 10(41.7) 20(6.3) χ2=34.484 <0.001
尿白细胞≥100个/μL[例(%)] 18(75.0) 140(44.4) χ2=7.837 0.004
手术时间[min，M(Q)] 98(86，149) 100(80，127) Z=-0.586 0.558
白细胞计数[109/L，M(Q)] 6.59(4.95，8.19) 6.27(5.55，7.48) Z=-0.206 0.837
淋巴细胞计数[109/L，M(Q)] 1.72(1.42，2.31) 1.83(1.43，2.20) Z=-0.333 0.739
单核细胞计数[109/L，M(Q)] 0.41(0.34，0.51) 0.39(0.31，0.48) Z=-1.183 0.237
中性粒细胞计数[109/L，M(Q)] 3.95(2.65，5.40) 3.92(3.28，4.77) Z=-0.012 0.991
嗜酸性粒细胞计数[109/L，M(Q)] 0.09(0.05，0.19) 0.12(0.07，0.21) Z=-1.653 0.098
嗜碱性粒细胞计数[109/L，M(Q)] 0.03(0.02，0.04) 0.03(0.02，0.04) Z=-0.235 0.814
红细胞计数[1012/L，M(Q)] 4.46(4.20，4.70) 4.63(4.27，4.97) Z=-1.234 0.217
血红蛋白[g/L，M(Q)] 131.5(125.3，145.0) 142(130，152) Z=-2.040 0.041
红细胞分布宽度[%，M(Q)] 12.9(12.5，13.3) 12.7(12.3，13.2) Z=-0.738 0.461
血小板计数[109/L，M(Q)] 240.5(184.3，297.5) 219(181，262) Z=-1.196 0.232
丙氨酸氨基转移酶[U/L，M(Q)] 16.7(8.2，24.0) 18.8(13.3，28.8) Z=-1.540 0.124
天门冬氨酸氨基转移酶[U/L，M(Q)] 20.8(14.1，22.6) 19.3(16.3，23.9) Z=-0.279 0.781
血总蛋白[g/L，M(Q)] 67.9(63.9，71.6) 71.2(65.1，74.6) Z=-1.974 0.048
血白蛋白[g/L，M(Q)] 41.7(39.3，44.0) 41.8(38.4，44.7) Z=-0.347 0.729
血球蛋白[g/L，M(Q)] 26.1(23.6，28.7) 28.4(26.1，33.6) Z=-3.069 0.002
血前白蛋白[mg/L，M(Q)] 235.8(197.7，279.3) 270.6(230.1，304.9) Z=-2.563 0.010
血肌酐[μmol/L，M(Q)] 66.4(54.7，79.8) 71.2(60.5，84.7) Z=-1.574 0.116
血尿酸[μmol/L，M(Q)] 302.5(270.3，413.3) 371.9(305.4，437.6) Z=-1.546 0.122
总胆固醇[mmol/L，M(Q)] 4.32(3.86，5.32) 4.48(4.01，5.14) Z=-0.509 0.611
甘油三酯[mmol/L，M(Q)] 1.27(1.03，1.98) 1.43(1.06，2.20) Z=-0.440 0.660
高密度脂蛋白胆固醇[mmol/L，M(Q)] 1.05(0.89，1.28) 1.02(0.90，1.21) Z=-0.048 0.962
低密度脂蛋白胆固醇[mmol/L，M(Q)] 2.46(2.08，3.13) 2.69(2.18，3.24) Z=-0.821 0.412
超敏C反应蛋白[mg/L，M(Q)] 2.35(1.14，8.12) 1.83(0.83，4.16) Z=-1.091 0.275
表2 脓毒症与非脓毒症患者单因素及多因素Logistic回归分析
项目 单因素分析 多因素分析
OR 95%CI P值 OR 95%CI P值
年龄≥52岁 0.921 0.401~2.112 0.845      
性别（男=1，女=0） 0.218 0.088~0.541 0.001 0.341 0.082~1.414 0.138
体质量指数≥24.6 kg/m2 1.409 0.608~3.267 0.424      
结石负荷≥363 mm2 3.097 1.198~8.007 0.020 1.948 0.657~5.777 0.229
术前疑似感染性结石 7.375 2.818~19.299 <0.001 5.589 1.659~18.834 0.005
多发结石 1.496 0.636~3.519 0.356      
肾积水 1.289 0.556~2.989 0.554      
高血压 0.556 0.215~1.441 0.227      
糖尿病 1.600 0.517~4.951 0.415      
既往泌尿系手术史 1.375 0.597~3.168 0.455      
尿细菌培养阳性 1.923 0.789~4.690 0.151      
尿亚硝酸盐阳性 10.536 4.160~26.685 <0.001 5.312 1.802~15.662 0.002
尿白细胞≥100个/μL 3.750 1.450~9.699 0.006 1.881 0.643~5.506 0.249
手术时间≥100 min 0.625 0.270~1.449 0.273      
白细胞计数≥6.29×109/L 1.267 0.551~2.915 0.577      
淋巴细胞计数≥1.82×109/L 0.852 0.370~1.958 0.705      
单核细胞计数≥0.39×109/L 1.205 0.524~2.770 0.661      
中性粒细胞计数≥3.93×109/L 1.189 0.517~2.735 0.683      
嗜酸性粒细胞计数≥0.12×109/L 0.692 0.298~1.604 0.391      
嗜碱性粒细胞计数≥0.03×109/L 1.086 0.474~2.491 0.845      
红细胞计数≥4.59×109/L 0.466 0.194~1.121 0.088      
血红蛋白≥141 g/L 0.384 0.155~0.952 0.039 1.335 0.328~5.440 0.687
红细胞分布宽度≥12.8% 1.334 0.580~3.067 0.498      
血小板计数≥221×109/L 1.699 0.722~3.996 0.225      
丙氨酸氨基转移酶≥18.6 U/L 0.701 0.302~1.625 0.407      
天门冬氨酸氨基转移酶≥19.4 U/L 1.236 0.537~2.841 0.619      
血总蛋白≥68.1 g/L 0.395 0.164~0.950 0.038 0.245 0.071~0.846 0.026
血白蛋白≥41.7 g/L 1.072 0.468~2.460 0.869      
血球蛋白≥26.3 g/L 0.356 0.144~0.882 0.026 0.991 0.300~3.270 0.988
血前白蛋白≥268.7 mg/L 0.389 0.157~0.964 0.041 0.404 0.133~1.228 0.110
血肌酐≥70.8 μmol/L 2.100 0.905~4.875 0.084      
血尿酸≥369.4 μmol/L 0.574 0.244~1.350 0.203      
总胆固醇≥4.47 mmol/L 0.830 0.361~1.909 0.661      
甘油三酯≥1.4 mmol/L 0.896 0.390~2.060 0.796      
高密度脂蛋白胆固醇≥1.02 mmol/L 1.130 0.492~2.600 0.773      
低密度脂蛋白胆固醇≥2.68 mmol/L 0.581 0.247~1.367 0.214      
超敏C反应蛋白≥1.83 mg/L 0.994 0.433~2.279 0.988      
图1 LASSO回归选择预测PCNL术后脓毒症的肾结石影像组学特征过程及结果注：a示随着正则化参数λ值增加，各影像组学特征系数的变化轨迹，每条曲线代表一个影像组学特征；b为交叉验证均方误差图，显示不同λ值对应的交叉验证均方误差，虚线表示最小误差对应的λ值，用于最终特征选择；c为最终选择的8个影像组学特征权重分布，显示LASSO回归筛选出的8个特征及其对应的回归系数权重
图2 不同机器学习模型预测PCNL术后脓毒症的ROC曲线比较注：a为基于3个临床特征的8种机器学习模型在测试集中的ROC曲线比较，ExtraTrees模型表现最佳（AUC=0.743）；b为基于8个CT影像组学特征的8种机器学习模型在测试集中的ROC曲线比较，KNN模型表现最佳（AUC=0.878）；c为临床模型、CT影像组学模型和联合模型的ROC曲线比较，联合模型AUC达0.898，优于单一模型；LR=逻辑回归，SVM=支持向量机，KNN=K近邻算法，RF=随机森林，ExtraTrees=极端随机树，XGBoost=极端梯度提升，LightGBM=轻量梯度提升机，MLP=多层感知机
图3 PCNL术后脓毒症预测联合模型注：该列线图整合了3个临床特征（感染性结石、尿亚硝酸盐、血总蛋白）和CT影像组学特征，用于预测PCNL术后脓毒症发生率
图4 不同预测PCNL术后脓毒症模型校准曲线注：a为训练集校准曲线，显示预测概率与实际发生概率的一致性，对角线为完全校准线，实际曲线越接近对角线表示校准效果越好；b为测试集校准曲线
图5 不同预测PCNL术后脓毒症模型决策曲线注：a为训练集决策曲线显示不同阈值概率下模型的净效益，联合模型（绿线）在大部分阈值范围内优于临床模型（蓝线）和影像组学模型（橙线）；b为测试集决策曲线，证实联合模型具有更高的临床应用价值
[1]
Hill AJ, Basourakos SP, Lewicki P, et al. Incidence of kidney stones in the United States: the continuous national health and nutrition examination survey[J]. J Urol, 2022, 207(4): 851-856. DOI: 10.1097/ju.0000000000002331.
[2]
苏潇哲, 赵延春, 杨嗣星. 1990—2019年中国尿路结石疾病负担和变化趋势分析[J]. 临床泌尿外科杂志, 2023, 38(8): 627-631. DOI: 10.13201/j.issn.1001-1420.2023.08.013.
[3]
郭宝玉, 俞俊杰, 於江泉, 等. 经皮肾镜术后并发脓毒症/脓毒性休克的研究进展[J/OL]. 中华腔镜泌尿外科杂志(电子版), 2019, 13(6): 424-428.
[4]
Hao X, Wang X, Wei H, et al. Development and validation of the prediction model of sepsis in patients after percutaneous nephrolithotomy and sepsis progresses to septic shock[J]. J Endourol, 2023, 37(4): 377-386. DOI: 10.1089/end.2022.0384.
[5]
Arthur A, Orton MR, Emsley R, et al. A CT-based radiomics classification model for the prediction of histological type and tumour grade in retroperitoneal sarcoma (RADSARC-R): a retrospective multicohort analysis[J]. Lancet Oncol, 2023, 24(11): 1277-1286. DOI: 10.1016/S1470-2045(23)00462-X.
[6]
Meng Y, Yang Y, Hu M, et al. Artificial intelligence-based radiomics in bone tumors: Technical advances and clinical application[J]. Semin Cancer Biol, 2023, 95: 75-87. DOI: 10.1016/j.semcancer.2023.07.003.
[7]
Guiot J, Vaidyanathan A, Deprez L, et al. A review in radiomics: Making personalized medicine a reality via routine imaging[J]. Med Res Rev, 2022, 42(1): 426-440. DOI: 10.1002/med.21846.
[8]
Ripa F, Cerrato C, Tandoğdu Z, et al. Clinical significance of stone culture during endourological procedures in predicting post-operative urinary sepsis: should it be a standard of care-evidence from a systematic review and meta-analysis from EAU section of Urolithiasis (EULIS)[J]. World J Urol, 2024, 42(1): 614. DOI: 10.1007/s00345-024-05319-0.
[9]
Singer M, Deutschman CS, Seymour CW, et al. The third international consensus definitions for sepsis and septic shock (sepsis-3)[J]. JAMA, 2016, 315(8): 801-810. DOI: 10.1001/jama.2016.0287.
[10]
徐宏博, 魏雪栋, 胡林昆, 等. 快速序贯器官衰竭评分对经皮肾镜取石术后脓毒症休克的预测价值[J]. 中华泌尿外科杂志, 2021, 42(5): 332-338. DOI: 10.3760/cma.j.cn112330-20200420-00312.
[11]
Yang Z, Lin D, Hong Y, et al. The effect of preoperative urine culture and bacterial species on infection after percutaneous nephrolithotomy for patients with upper urinary tract stones[J]. Sci Rep, 2022, 12(1): 4833. DOI: 10.1038/s41598-022-08913-7.
[12]
Qiu Z, Zhan S, Song Y, et al. Construction and validation of the nomogram predictive model for post-percutaneous nephrolithotomy urinary sepsis[J]. World J Urol, 2024, 42(1): 135. DOI: 10.1007/s00345-024-04828-2.
[13]
Wang L, Li D, He W, et al. Development and validation of a predictive model for post-percutaneous nephrolithotomy urinary sepsis: a multicenter retrospective study[J]. Minerva Urol Nephrol, 2024, 76(3): 357-366. DOI: 10.23736/S2724-6051.23.05396-X.
[14]
Yang ZY, Yang F. Nutritional status of patients with gastrointestinal cancers and analysis of factors for postoperative infections[J]. BMC Cancer, 2024, 24(1): 1389. DOI: 10.1186/s12885-024-13093-w.
[15]
刘蓬然, 霍彤彤, 陆林, 等. 人工智能在医学中的应用现状与展望[J]. 中华医学杂志, 2021, 101(44): 3677-3683. DOI: 10.3760/cma.j.cn112137-20210313-00628.
[16]
Hertel A, Froelich MF, Overhoff D, et al. Radiomics-driven spectral profiling of six kidney stone types with monoenergetic CT reconstructions in photon-counting CT[J]. Eur Radiol, 2025, 35(6): 3120-3130. DOI: 10.1007/s00330-024-11262-w.
[17]
Xun Y, Chen M, Liang P, et al. A novel clinical-radiomics model pre-operatively predicted the stone-free rate of flexible ureteroscopy strategy in kidney stone patients[J]. Front Med, 2020, 7: 576925. DOI: 10.3389/fmed.2020.576925.
[18]
Lu J, Zhu K, Yang N, et al. Radiomics and clinical features for distinguishing kidney stone-associated urinary tract infection: a comprehensive analysis of machine learning classification[J]. Open Forum Infect Dis, 2024, 11(10): ofae581. DOI: 10.1093/ofid/ofae581.
[1] 邓鸿飞, 冯苛, 王征, 蒋文凯, 张晓婷. 氢溴酸山莨菪碱（654-1）联合血液净化治疗脓毒症休克患者短期疗效[J/OL]. 中华实验和临床感染病杂志(电子版), 2026, 20(01): 31-38.
[2] 黄楚曦, 吴卓. 人工智能在膀胱癌影像中的应用进展[J/OL]. 中华腔镜泌尿外科杂志(电子版), 2026, 20(03): 241-247.
[3] 刘丰博, 杨德林, 夏成兴, 杨晓芳, 余航, 张平新, 李志鹏, 王留芳, 李菊, 叶春伟. 集落刺激因子-1及1，25（OH）2D3对巨噬细胞增殖极化的作用及其对草酸钙肾结石形成的影响[J/OL]. 中华腔镜泌尿外科杂志(电子版), 2026, 20(02): 202-208.
[4] 邓玉飞, 王志鑫, 娄珂, 张林轩, 马桂春, 港措. 影像组学在肝癌精准诊断、疗效评估及治疗方案决策优化中应用[J/OL]. 中华肝脏外科手术学电子杂志, 2026, 15(02): 172-180.
[5] 邢颖, 王峰. 基于机器学习构建肝切除术后肝衰竭预测列线图模型及其预测价值[J/OL]. 中华肝脏外科手术学电子杂志, 2026, 15(02): 190-196.
[6] 王小振, 陈灿辉, 唐善华, 代浩嘉, 丰扬舸, 王恺, 李清平, 李川江. 基于不同机器学习技术构建肝移植术后早期严重并发症预测模型和效能比较[J/OL]. 中华肝脏外科手术学电子杂志, 2026, 15(02): 197-204.
[7] 戴宗伯, 张城硕, 郭庭维, 何知远, 赵昊宇, 张宇慈, 张佳林. 基于MRI影像组学机器学习构建肝细胞癌微血管侵犯预测模型[J/OL]. 中华肝脏外科手术学电子杂志, 2026, 15(01): 36-44.
[8] 何泰霖, 王峻峰, 田林云, 王罡, 杨超, 王海峰. 基于CT影像组学构建肝细胞癌微血管侵犯预测模型[J/OL]. 中华肝脏外科手术学电子杂志, 2026, 15(01): 45-52.
[9] 钟文卿, 韩冰. 基于术前CT影像数据构建肝癌患者生存期Nomogram预测模型[J/OL]. 中华肝脏外科手术学电子杂志, 2026, 15(01): 53-58.
[10] 刘郁芳, 赵青. 直肠癌MRI影像学评估：从精准分期到预后预测的研究进展与展望[J/OL]. 中华结直肠疾病电子杂志, 2026, 15(01): 31-36.
[11] 许萍萍, 宋丹阳, 李笑, 李辉, 王霞. 降钙素原与氨基末端B型利钠肽原联合评分对甲泼尼龙联合集成化血液净化治疗儿童严重脓毒症效果的分层价值[J/OL]. 中华肾病研究电子杂志, 2026, 15(02): 93-99.
[12] 史楠, 于瑞璇, 谢剑锋, 邱海波. 慢病脓毒症的疾病负担及研究现状[J/OL]. 中华重症医学电子杂志, 2026, 12(01): 7-14.
[13] 席寅, 高妍纯, 王炜林, 徐永昊. 慢病脓毒症的液体复苏：器官储备功能导向的个体化策略[J/OL]. 中华重症医学电子杂志, 2026, 12(01): 15-20.
[14] 陈芊慧, 崔志刚, 孙晋赫, 丁仁彧. 脓毒症相关重度血小板减少症患者的多因素预后分析及预后模型构建[J/OL]. 中华重症医学电子杂志, 2026, 12(01): 53-66.
[15] 陈小坤, 杜顺达. 影像组学在肝细胞癌中的应用进展及挑战[J/OL]. 中华消化病与影像杂志(电子版), 2026, 16(02): 97-100.
阅读次数
全文


摘要

版权所有 中华医学会 中华医学电子音像出版社有限责任公司  京ICP备14006079号-1  网络出版服务许可证：(署)网出证(京)字第075号

AI
小
编
AI小编
你好！我是《中华医学电子期刊资源库》AI小编，有什么可以帮您的吗？