Comparison of artificial neural network, extreme gradient boosting and logistic regression for predicting intraoperative transfusion in repeat cesarean delivery

Jie LI; Guangyou DUAN; Yi ZENG; Zhenxin DUAN; Zhuoxi WU; Guiying YANG; Hong LI

doi:10.16016/j.1000-5404.201907118

您当前的位置：

首页 >

文章列表页 >

Comparison of artificial neural network, extreme gradient boosting and logistic regression for predicting intraoperative transfusion in repeat cesarean delivery

Clinical Medicine | 更新时间：2026-01-09

- Comparison of artificial neural network, extreme gradient boosting and logistic regression for predicting intraoperative transfusion in repeat cesarean delivery
- Journal of Third Military Medical University Vol. 41, Issue 24, Pages: 2430-2437(2019)
- 作者机构：
  
  400037 重庆，陆军军医大学(第三军医大学)第二附属医院麻醉科
- 作者简介：
- 基金信息：
  
  the National Key Research and Development Program(2018YFC0117200);the Special Fund for Military Clinical Medicine Innovation of Second Affiliated Hospital, Army Medical University(2018JSLC0015);the Clinical Technology Innovation Project of Army Medical University(CX2019LC114)
- DOI：10.16016/j.1000-5404.201907118
  CLC： R457.1;R614;R719.8
- Received：15 July 2019，
  
  Revised：2019-09-20，
  
  Published：30 December 2019
- 稿件说明：
移动端阅览
Jie LI, Guangyou DUAN, Yi ZENG, et al. Comparison of artificial neural network, extreme gradient boosting and logistic regression for predicting intraoperative transfusion in repeat cesarean delivery[J]. Journal of Third Military Medical University, 2019, 41(24): 2430-2437.
DOI：

Jie LI, Guangyou DUAN, Yi ZENG, et al. Comparison of artificial neural network, extreme gradient boosting and logistic regression for predicting intraoperative transfusion in repeat cesarean delivery[J]. Journal of Third Military Medical University, 2019, 41(24): 2430-2437. DOI： 10.16016/j.1000-5404.201907118.

摘要

目的

通过大样本临床数据构建再次剖宫产术中输血的预测模型，分析比较人工神经网络(artificial neural network，ANN)、极端梯度提升(extreme gradient boosting，XGB)和Logistic回归3种机器学习算法的预测效果。

方法

通过医院病历系统，收集我院2015年10月至2017年10月符合纳入标准的再次剖宫产产妇2 525例，详细录入产妇术前、术中各项指标。将可能有临床意义的变量纳入预测模型的构建，分别采用Logistic回归、XGB和ANN 3种机器学习算法构建术中输血预测模型。计算并比较3个模型的ROC曲线下面积(AUROC)、精确度、召回率和F1值4个指标。

结果

研究的有效样本共2 525例，其中332例(13.1%)进行了术中输血。最终的模型得到5个最重要的预测因素为:术前Hb、手术时间、宫缩乏力、前置胎盘和ASA分级。Logistic回归、XGB和ANN 3种算法的AUROC依次为0.960、0.959、0.956，同时通过F1值、精确度、召回率3个指标的比较发现3种算法差别很小。为进一步比较预测效果，再次在训练样本和测试样本上进行预测验证，发现XGB的AUROC为0.904和0.886，高于Logistic回归的0.868和0.878，以及ANN的0.882和0.884，同时XGB的精确度、召回率和F1 3个指标均略高于Logistic回归和ANN。

结论

术前Hb、手术时间、前置胎盘等指标可用于预测再次剖宫产术中输血。Logistic回归、XGB和ANN这3种机器学习算法均可用于剖宫产术中输血的预测，但XGB的预测效果比Logistic回归和ANN更准确。

Abstract

Objective

To construct prediction models for intraoperative transfusion in repeat cesarean delivery using logistic regression

extreme gradient boosting (XGB) and artificial neural network (ANN) based on the clinical data with a large sample size and compare their prediction performance.

Methods

From October

2015 to October

2017

a total of 2 525 women underwent repeat cesarean delivery in our hospital. We analyzed the preoperative and intraoperative data of the women

and the variables with possible clinical significance were selected to construct the prediction models for intraoperative transfusion using logistic regression

XGB and ANN. The area under the receiver operating characteristic curve (AUROC)

precision

recall (sensitivity) and F1 score of the 3 models were compared.

Results

Among the 2 525 women included in this study

332 (13.1%) received intraoperative transfusion in repeat cesarean delivery. The final model was constructed based on 5 important predictors

including preoperative hemoglobin

operation time

uterine atony

placenta previa and ASA classification. The AUROC of the algorithms of logistic regression

XGB and ANN were 0.958

0.959

and 0.956

respectively

and their F1

precision and recall did not differ significantly. To compare their prediction performance

further predictive verification was carried out on training samples and test samples

for which XGB yielded AUROC of 0.904 and 0.886

respectively

significantly higher than those of logistic regression (0.868 and 0.878) and ANN (0.882 and 0.884). The precision

recall and F1 of XGB were also better than those of logistic regression and ANN.

Conclusion

Preoperative hemoglobin

operation time

uterine atony

placenta previa and ASA classification are predictors of intraoperative transfusion in repeat cesarean delivery. Logistic regression

XGB and ANN can all be used for predicting intraoperative transfusion in repeat cesarean delivery

but XGB has higher prediction accuracy than logistic regression and ANN.

关键词

Keywords

references

SAAD A, COSTANTINE M M. Obstetric hemorrhage: recent advances[J]. Clin Obstet Gynecol, 2014, 57(4): 791-796. DOI: 10.1097/GRF.0000000000000062.

PEARSON G A, MACKENZIE I Z. Blood loss and blood transfusion at Caesarean section: a prospective observational study covering 30years[J]. Eur J Obstet Gynecol Reprod Biol, 2014, 181: 72-77. DOI: 10.1016/j.ejogrb.2014.06.025.

CHUA S C, JOUNG S J, AZIZ R. Incidence and risk factors predicting blood transfusion in Caesarean section[J]. Aust N Z J Obstet Gynaecol, 2009, 49(5): 490-493. DOI:10.1111/j.1479-828X.2009.01042.x.

GOUNDAN A, KALRA J K, RAVEENDRAN A, et al . Descriptive study of blood transfusion practices in women undergoing cesarean delivery[J ] . J Obstet Gynaecol Res, 2011, 37(10): 1277-1282. DOI: 10.1111/j.1447-0756.2010.01511.x.

曾义, 李杰, 段光友, 等.再次剖宫产产后出血危险因素分析: 2442例产妇的回顾性研究[J].第三军医大学学报, 2019, 41(3): 265-269. DOI: 10.16016/j.1000-5404.201809015.

ZENG Y, LI J, DUAN G Y, et al . Risk factors for postpartum hemorrhage in repeat cesarean delivery: a retrospective study of 2 442 women[J ] . J Third Mil Med Univ, 2019, 41(3): 265-269. DOI: 10.16016/j.1000-5404.201809015.

ROUSE D J, MACPHERSON C, LANDON M, et al . Blood transfusion and cesareandelivery[J ] . Obstet Gynecol, 2006, 108(4): 891-897. DOI: 10.1097/01.AOG.0000236547.35234.8c.

SPIEGELMAN J, MOURAD M, MELKA S, et al . Risk factors for blood transfusion in patients unde rgoing high-order Cesarean delivery[J ] . Transfusion, 2017, 57(11): 2752-2757. DOI: 10.1111/trf.14274.

KOUROU K, EXARCHOS T P, EXARCHOS K P, et al . Machine learning applications in cancer prognosis and prediction[J ] . Comput Struct Biotechnol J, 2015, 13: 8-17. DOI: 10.1016/j.csbj.2014.11.005.

ZACHARAKI E I, KANAS V G, DAVATZIKOS C. Investigating machine learning techniques for MRI-based classification of brainneoplasms[J]. Int J Comput Assist Radiol Surg, 2011, 6(6): 821-828. DOI: 10.1007/s11548-011-0559-3.

KENDALE S, KULKARNI P, ROSENBERG A D, et al . Supervised machine-learning predictive analytics for prediction ofpostinduction hypotension[J ] . Anesthesiology, 2018, 129(4): 675-688. DOI: 10.1097/ALN.0000000000 002374.

LEE C K, HOFER I, GABEL E, et al . Development and validation of a deep neural network model for prediction of postoperative in-hospitalmortality[J ] . Anesthesiology, 2018, 129(4): 649-662. DOI: 10.1097/ALN.0000000 000002186.

LECUN Y, BENGIO Y, HINTON G. Deeplearning[J]. Nature, 2015, 521(7553): 436-444. DOI: 10.1038/nature14539.

BAO Y R, XU C L, QU X X, et al . Risk factors for transfusion in cesarean section deliveries at a tertiary hospital[J ] . Transfusion, 2016, 56(8): 2062-2068. DOI: 10.1111/trf.13671.

BUTWICK A J, RAMACHANDRAN B, HEGDE P, et al . Risk factors for severe postpartum hemorrhage after cesarean delivery: case-controlstudies[J ] . Anesth Analg, 2017, 125(2): 523-532. DOI: 10.1213/ANE.0000000000001962.

曹文哲, 应俊, 张亚慧, 等.基于机器学习算法的前列腺癌诊断模型研究[J].中国医疗设备, 2016, 31(4): 30-35.

CAO W Z, YING J, ZHANG Y H, et al . Diagnostic model research of prostate cancer based on machine learning algorithm[J ] . China Med Devices, 2016, 31(4): 30-35.

GIZZO S, ZAMBON A, SACCARDI C, et al . Effective anatomical and functional status of the lower uterine segment at term: estimating the risk of uterine dehiscence byultrasound[J ] . Fertil Steril, 2013, 99(2): 496-501. DOI: 10.1016/j.fertnstert.2012.10.019.

SHERIDAN R P, WANG W M, LIAW A, et al . Extreme gradient boosting as a method for quantitative structure-activity relationships[J ] . J Chem Inf Model, 2016, 56(12): 2353-2360. DOI: 10.1021/acs.jcim.6b00591.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Biomarkers for hemorrhagic shock at high-altitude based on metabolomics and machine learning: Identification and validation of D-inositol-4-phosphate, phosalone and methionine

Risk factors for postoperative respiratory failure in patients undergoing cardiovascular surgery and construction of a prediction model

Development of a postoperative recurrence prediction model for stage Ⅰ non-small cell lung cancer patients using multimodal data based on machine learning

Construction of machine learning-based prognostic model of centrosome amplification-related genes for esophageal squamous cell carcinoma

Construction and validation of a prediction model for central lymph node metastasis of papillary thyroid carcinoma based on contrast-enhanced venous phase CT radiomics

Related Author

ZHOU Yuanqun

XIANG Xinming

OUYANG Xingnan

ZHANG Jie

ZHANG Zisen

LIU Liangming

LI Tao

ZHOU Fuchao

Related Institution

National Key Laboratory of Trauma and Chemical Poisoning， Department of War Trauma， Shock and Blood Transfusion， Daping Hospital， Army Medical University（Third Military Medical University）

Department of Anesthesiology， First Affiliated Hospital， Army Medical University （Third Military Medical University）

Jinfeng Laboratory

Yu-Yue Pathology Scientific Research Center

Department of Digital Medicine， Faculty of Biomedical Engineering and Medical Imaging， Army Medical University （Third Military Medical University）

⁰