基于改进的机器学习模型对重症急性胰腺炎诊断的早期预测

李龙; 尹梁宇; 种菲菲; 童宁; 黎娜; 刘洁; 余相江; 王耀丽; 许红霞

doi:10.16016/j.2097-0927.202309150

您当前的位置：

首页 >

文章列表页 >

基于改进的机器学习模型对重症急性胰腺炎诊断的早期预测

临床医学 | 更新时间：2026-01-09

- 基于改进的机器学习模型对重症急性胰腺炎诊断的早期预测
- Early prediction of severe acute pancreatitis based on improved machine learning models
- 陆军军医大学学报 2024年46卷第7期页码：753-759
- 作者机构：
  
  1.400042 重庆, 陆军特色医学中心临床营养科
  2.625000 四川雅安, 联勤保障部队第九四五医院急诊与重症医学科
  3.400042 重庆, 陆军特色医学中心重症医学科
- 作者简介：
  
  [许红霞，E-mail：hx_xu2015@163.comXU Hongxia, E-mail: hx_xu2015@163.com
- 基金信息：
  
  陆军特色医学中心军事医学前沿创新能力培养计划(2020ljts009)
- DOI：10.16016/j.2097-0927.202309150
  中图分类号： R181.32;R319;R576.04
- 收稿：2023-09-28，
  
  修回：2023-12-15，
  
  纸质出版：2024-04-15
- 稿件说明：
移动端阅览
李龙, 尹梁宇, 种菲菲, 等. 基于改进的机器学习模型对重症急性胰腺炎诊断的早期预测[J]. 陆军军医大学学报, 2024,46(7):753-759.

Long LI, Liangyu YIN, Feifei CHONG, et al. Early prediction of severe acute pancreatitis based on improved machine learning models[J]. Journal of Army Medical University, 2024, 46(7): 753-759.
李龙, 尹梁宇, 种菲菲, 等. 基于改进的机器学习模型对重症急性胰腺炎诊断的早期预测[J]. 陆军军医大学学报, 2024,46(7):753-759. DOI： 10.16016/j.2097-0927.202309150.

Long LI, Liangyu YIN, Feifei CHONG, et al. Early prediction of severe acute pancreatitis based on improved machine learning models[J]. Journal of Army Medical University, 2024, 46(7): 753-759. DOI： 10.16016/j.2097-0927.202309150.

摘要

目的

基于改进的机器学习模型建立重症急性胰腺炎诊断的早期预测模型

并分析其临床价值。

方法

纳入2014年1月至2023年8月陆军特色医学中心消化内科、肝胆外科以及联勤保障部队第九四五医院急诊与重症医学科收治的352例急性胰腺炎患者

根据病情严重程度将其分为重症组(

=88)和非重症组(

=264)

开展病例对照研究。利用RUSBoost模型以及改进的阿基米德优化算法

分析入院48 h内的39项常规实验室生化指标

帮助构建重症急性胰腺炎早期诊断预测模型

同步完成特征筛选和超参数优化

并利用ReliefF算法特征重要性排序和Logistic多因素分析

对筛选出的特征进行价值分析。

结果

在训练集上

改进机器学习模型的曲线下面积(area under curve

AUC)为0.922;在测试集上

改进机器学习模型的AUC达到了0.888。基于改进机器学习模型筛选出的预测重症急性胰腺炎发生的4个关键特征分别为C反应蛋白、血氯、血镁、纤维蛋白原水平

与ReliefF算法特征重要性排序和Logistic多因素分析结果相吻合。

结论

应用改进机器学习模型分析实验室检查结果

可帮助临床早期预测重症急性胰腺炎的发生。

Abstract

Objective

To establish an early prediction model for the diagnosis of severe acute pancreatitis based on the improved machine learning models

and to analyze its clinical value.

Methods

A case-control study was conducted on 352 patients with acute pancre

atitis admitted to the Gastroenterology and Hepatobiliary Surgery Departments of the Army Medical Center of PLA and Emergency and Critical Care Medicine Department of No.945 Hospital of Joint Logistics Support Force of PLA from January 2014 to August 2023.According to the severity of the disease

the patients were divided into the severe group (

=88) and the non-severe group (

=264).The RUSBoost model and improved Archimead optimization algorithm was used to analyze 39 routine laboratory biochemical indicators within 48 h after admission to construct an early diagnosis and prediction model for severe acute pancreatitis.The task of feature screening and hyperparameter optimization was completed simultaneously.The ReliefF algorithm feature importance rank and multivariate logistic analysis were used to analyze the value of the selected features.

Results

In the training set

the area under curve (AUC) of the improved machine learning model was 0.922.In the testing set

the AUC of the improved machine learning model reached 0.888.The 4 key features of predicting severe acute pancreatitis based on the improved Archimedes optimization algorithm were C-reactive protein

blood chlorine

blood magnesium and fibrinogen level

which were consistent with the results of ReliefF algorithm feature importance ranking and multivariate logistic analysis.

Conclusion

The application of improved machine learning model analyzing the laboratory examination results can help to early predict the occurrence of severe acute pancreatitis.

关键词

Keywords

references

ZHOU Q, XIANG H, LIU H, et al . Emodin alleviates intestinal barrier dysfunction by inhibiting apoptosis and regulating the immune response in severe acute pancreatitis[J ] . Pancreas, 2021, 50(8): 1202-1211. DOI: 10.1097/MPA.0000000000001894.

MEDEROS M A, REBERH A, GIRGIS M D. Acute pancreatitis[J]. JAMA, 2021, 325(4): 382. DOI: 10.1001/jama.2020.20317.

毛恩强. 改善重症急性胰腺炎预后的"瓶颈" 与思考[J]. 国际外科学杂志, 2020, 47(5): 299-301. DOI: 10.3760/cma.j.cn115396-20200426-00116.

MAO E Q. "Bottleneck" and thinking of improving the prognosis of severe acute pancreatitis[J]. Int J Surg, 2020, 47(5): 299-301. DOI: 10.3760/cma.j.cn115396-20200426-00116.

GUPTA M, LITI B, BARRETT C, et al . Prevention and management of hypertriglyceridemia-induced acute pancreatitis during pregnancy: a systematic review[J ] . Am J Med, 2022, 135(6): 709-714. DOI: 10.1016/j.amjmed.2021.12.006.

GREENER J G, KANDATHIL S M, MOFFAT L, et al . A guide to machine learning for biologists[J ] . Nat Rev Mol C ell Biol, 2022, 23(1): 40-55. DOI: 10.1038/s41580-021-00407-0.

范海云, 陈基明, 陈亮亮, 等. 基于CT影像组学特征的机器学习模型预测早期急性胰腺炎进展的价值[J]. 皖南医学院学报, 2022, 41(3): 256-259. DOI: 10.3969/j.issn.1002-0217.2022.03.015.

FAN H Y, CHEN J M, CHEN L L, et al . Predicting progression of early acute pancreatitis by machine learning model based on CT radiomics[J ] . Acta Acad Med Wannan, 2022, 41(3): 256-259. DOI: 10.3969/j.issn.1002-0217.2022.03.015.

CHOI R Y, COYNER A S, KALPATHY-CRAMER J, et al . Introduction to machine learning, neural networks, and deep learning[J ] . Transl Vis Sci Technol, 2020, 9(2): 14. DOI: 10.1167/tvst.9.2.14.

杨磊, 曹锋, 李非. 急性胰腺炎《修订版亚特兰大分类标准》的应用评价[J]. 中华肝胆外科杂志, 2015, 21(6): 365-368. DOI: 10.3760/cma.j.issn.1007-8118.2015.06.002.

YANG L, CAO F, LI F. Validation of the revised Atlanta Classification and the determinant-based classification for severity of acute pancreatitis[J ] . Chin J Hepatobil Surg, 2015, 21(6): 365-368. DOI: 10.3760/cma.j.issn.1007-8118.2015.06.002.

中华医学会消化病学分会胰腺疾病学组, 中华胰腺病杂志编辑委员会, 中华消化杂志编辑委员会. 中国急性胰腺炎诊治指南(2019年, 沈阳)[J]. 中华消化杂志, 2019, 39(11): 721-730. DOI: 10.3760/cma.j.issn.0254-1432.2019.11.001.

Pancreatic Disease Group of Digestive Disease Branch of Chinese Medical Association, Editorial Committee of Chinese Journal of Pancreatic Disease , Editorial Committee of Chinese Journal of Gastroenterology . Guidelines for diagnosis and treatment of acute pancreatitis in China (Shenyang, 2019)[J ] . Chin J Dig, 2019, 39(11): 721-730. DOI: 10.3760/cma.j.issn.0254-1432.2019.11.001..

安宏达, 孙备. 重症急性胰腺炎救治的难点与困惑[J]. 中华胰腺病杂志, 2020, 20(5): 325-327. DOI: 10.3760/cma.j.cn115667-20200721-00115.

AN H D, SUN B. Difficulties and perplexities on the treatment of severe acute pancreatitis[J]. Chin J Pancreatol, 2020, 20(5): 325-327. DOI: 10.3760/cma.j.cn115667-20200721-00115.

陈鑫龙, 叶凯, 周文策. 人工智能在胰腺疾病新型诊疗模式中的应用及进展[J]. 中国医学物理学杂志, 2022, 39(8): 1049-1056. DOI: 10.3969/j.issn.1005-202X.2022.08.023.

CHEN X L, YE K, ZHOU W C. Application of artificial intelligence in new diagnostic and therapeutic pattern of pancreatic diseases and its advances[J]. Chin J Med Phys, 2022, 39(8): 1049-1056. DOI: 10.3969/j.issn.1005-202X.2022.08.023.

丁莺, 周道扬, 何洋, 等. 基于机器学习的急性重症胰腺炎早期预测系统[J]. 中华急诊医学杂志, 2020, 29(10): 1343-1347. DOI: 10.3760/cma.j.issn.1671-0282.2020.10.013.

DING Y, ZHOU D Y, HE Y, et al . Early prediction system for acute severe pancreatit is based on machine learning[J ] . Chin J Emerg Med, 2020, 29(10): 1343-1347. DOI: 10.3760/cma.j.issn.1671-0282.2020.10.013.

季顾惟, 王科, 夏永祥, 等. 机器学习算法在早期肝细胞癌术后复发预测中的应用价值[J]. 中华外科杂志, 2021, 59(8): 679-685. DOI: 10.3760/cma.j.cn112139-20201026-00768.

JI G W, WANG K, XIA Y X, et al . Application value of machine learning algorithms for predicting recurrence after resection of early-stage hepatocellular carcinoma[J ] . Chin J Surg, 2021, 59(8): 679-685. DOI: 10.3760/cma.j.cn112139-20201026-00768.

范可欣, 朱鹏汇, 王云, 等. 基于机器学习算法建立胎母输血综合征预测模型[J]. 临床输血与检验, 2022, 24(4): 427-432. DOI: 10.3969/j.issn.1671-2587.2022.04.004.

FAN K X, ZHU P H, WANG Y, et al . The prediction model of fetomaternal hemorrhage was established on machine learning algorithm[J ] . J Clin Transfus Lab Med, 2022, 24(4): 427-432. DOI: 10.3969/j.issn.1671-2587.2022.04.004.

张福龙, 邱兆磊, 王振杰, 等. 肝素结合蛋白联合C反应蛋白在重症急性胰腺炎早期病情评估中的临床意义[J]. 中华全科医学, 2021,19(3): 368-370, 374. DOI: 10.16766/j.cnki.issn.1674-4152.001812.

ZHANG F L, QIU Z L, WANG Z J, et al . The clinical significance of heparin-binding protein combined with CRP in the early evaluation of severe acute pancreatitis[J ] . Chin J Gen Pract, 2021, 19(3): 368-370, 374. DOI: 10.16766/j.cnki.issn.1674-4152.001812.

STIRLING A D, MORAN N R, KELLY M E, et al . The predictive value of C-reactive protein (CRP) in acute pancreatitis—is interval change in CRP an additional indicator of severity?[J ] . HPB (Oxford), 2017, 19(10): 874-880. DOI: 10.1016/j.hpb.2017.06.001.

邓中波, 叶淳娟, 徐玉龙. 血清血小板平均体积CRP纤维蛋白原与急性胰腺炎患者病情程度和预后转归的关系分析[J]. 河北医学, 2022, 28(4): 670-675. DOI: 10.3969/j.issn.1006-6233.2022.04.028.

DENG Z B, YE C J, XU Y L. Analysis of the relationship of mean platelet volume C-reactive protein and fibrinogen with the severity and prognosis of acute pancreatitis[J]. Hebei Med, 2022, 28(4): 670-675. DOI: 10.3969/j.issn.1006-6233.2022.04.028.

崔沪宁, 梅超鹏, 崔梦巍, 等. 血氯和IL-6联合凝血指标对重症急性胰腺炎严重程度的预测价值[J]. 中华危重病急救医学, 2022, 34(12): 1301-1304. . DOI: 10.3760/cma.j.cn121430-20220916-00844.

CUI H N, MEI C P, CUI M W, et al . Predictive value of serum chlorine and interleukin-6 combined with coagulation indexes on severity of severe acute pancreatitis[J ] . Chin Crit Care Med, 2022, 34(12): 1301-1304. DOI: 10.3760/cma.j.cn121430-20220916-00844.

YU X Q, DENG H B, LIU Y, et al . Serum magnesium level as a predictor of acute kidney injury in patients with acute pancreatitis[J ] . World J Clin Cases, 2021, 9(35): 10899-109 08. DOI: 10.12998/wjcc.v9.i35.10899.

HUANG L Y, MA B W, HE F, et al . Electrocardiographic, cardiac enzymes, and magnesium in patients with severe acute pancreatitis[J ] . Gastroenterol Nurs, 2012, 35(4): 256-260. DOI: 10.1097/sga.0b013e31826092a6.

浏览量

497

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

住院期间ICU重症急性胰腺炎患者家属迁移应激水平变化及相关因素研究

重庆市某医院89例新型冠状病毒肺炎患者的临床特征分析

胎盘间充质干细胞移植对大鼠重症急性胰腺炎的保护作用

早期腹腔穿刺引流对重症急性胰腺炎大鼠肠道细菌移位的影响