Risk prediction model for postoperative cognitive dysfunction after total knee replacement based on Bayesian network algorithm

Danfeng LI; Haoyang LI; Kunhua ZHONG; Xin SHU; Yujie LI; Bin YI

doi:10.16016/j.2097-0927.202301054

您当前的位置：

首页 >

文章列表页 >

Risk prediction model for postoperative cognitive dysfunction after total knee replacement based on Bayesian network algorithm

Monographic Reports: Governance and artificial intelligence algorithms of big data during perioperative period | 更新时间：2026-01-09

- Risk prediction model for postoperative cognitive dysfunction after total knee replacement based on Bayesian network algorithm
- Journal of Army Medical University Vol. 45, Issue 8, Pages: 765-771(2023)
- 作者机构：
  
  1.400038 重庆, 陆军军医大学(第三军医大学)第一附属医院麻醉科
  2.400038 重庆, 陆军军医大学(第三军医大学)基础医学院学员五大队
  3.400714 重庆, 中国科学院重庆绿色智能技术研究院
- 作者简介：
- 基金信息：
  
  the National Key Research and Development Program(2018YFC0116702);the Special Project of Chongqing Technology Innovation and Application Research and Development in 2019(cstc2019jscx-msxmX0237)
- DOI：10.16016/j.2097-0927.202301054
  CLC： R319;R687.4;R749
- Received：01 January 2023，
  
  Revised：2023-02-16，
  
  Published：30 April 2023
- 稿件说明：
移动端阅览
Danfeng LI, Haoyang LI, Kunhua ZHONG, et al. Risk prediction model for postoperative cognitive dysfunction after total knee replacement based on Bayesian network algorithm[J]. Journal of Army Medical University, 2023, 45(8): 765-771.
DOI：

Danfeng LI, Haoyang LI, Kunhua ZHONG, et al. Risk prediction model for postoperative cognitive dysfunction after total knee replacement based on Bayesian network algorithm[J]. Journal of Army Medical University, 2023, 45(8): 765-771. DOI： 10.16016/j.2097-0927.202301054.

摘要

目的

利用贝叶斯网络(Bayesian network，BN)算法建立全膝关节置换术(total knee replacement，TKR)后认知功能障碍(postoperative cognitive dysfunction，POCD)的风险预测模型，并探讨模型的预测性能。

方法

采用病例-对照研究设计方案，选择陆军军医大学第一附属医院关节外科2017年1月至2021年12月行TKR的住院患者1 260例。入院主要诊断为左/右膝关节重度骨性关节炎，其中男性240例(19.0%)，女性1 020例(81.0%)；年龄23~79(66.73±8.46)岁；体质指数(BMI)(25.08±5.09)kg/m

。将术后(手术结束至出院期间)发生POCD的患者(71例)按照7∶3随机分为A1组(70%)和B1组(30%)，未发生POCD的患者(1 189例)按照7∶3随机分为A2组(70%)和B2组(30%)。A1组与A2组共同构成A组(训练数据集)，B1组与B2组则为B组(测试数据集)，其中A组用于模型训练，B组用于模型测试。筛选TKR相关围术期麻醉决策、病情转归及住院时间等36项指标作为节点，利用BN算法建立各节点的概率分布模型图，预测POCD发生风险的概率，以期住院时间最短，促进最大化患者康复。

结果

基于BN算法建立的预测TKR后POCD发生风险模型，训练集的受试者工作曲线下面积(area under the subject curve，AUC)值为0.966 1(95%

：0.954 1~0.978 4)，测试集AUC值为0.897 4(95%

：0.867 2~0.928 5)，准确性分别为96.43%(95%

：0.951 1~0.976 4)和93.44%(95%

：0.909 2~0.959 6)。

结论

基于BN算法建立的预测TKR后POCD发生风险的模型预测性能较好且准确率较高。

Abstract

Objective

To establish a prediction model of risk for postoperative cognitive dysfunction (POCD) after total knee replacement (TKR) by Bayesian network (BN) algorithm and investigate its predictive performance.

Methods

A case-control trial was conducted on 1 260 inpatients who underwent TKR from January 2017 to December 2021 in the Department of Joint Surgery of our hospital. Their main diagnosis of inclusion was severe osteoarthritis of left/right knee joint. They were 240 cases of male (19.0%) and 1 020 cases of female (81.0%)

at an average age of 66.73±8.46 (23~79) years and a mean body mass index (BMI) of 25.08±5.08 kg/m

. The POCD patients (

=71) after surgery (from the end of surgery to discharge) were randomly divided into A1 group and B1 group at a ratio of 7∶3

and those without POCD (1 189 cases) were also randomly divided into A2 group and B2 group at a same ratio. The patients from A1 and A2 groups were combined together as training set

and those out of B1 and B2 groups as test

set. Thirty-six indexes related to perioperative anesthesia decision

disease outcome and length of stay in TKR were selected as nodes

and the probability distribution model diagram of each node was established by using BN algorithm to predict the probability of risk for POCD

so as to minimize the length of stay and promote the maximum recovery of patients.

Results

The prediction model of risk for POCD after TKR was established based on BN algorithm. The area value under receiver operating characteristic curve (ROC-AUC) of the training set was 0.966 1 (95%

: 0.954 1~0.978 4)

and the ROC-AUC value of the test set was 0.897 4 (95%

: 0.867 2~0.928 5)

with an accuracy of 96.43% (95%

: 0.951 1~0.976 4) and 93.44% (95%

: 0.909 2~0.959 6)

respectively.

Conclusion

Our risk prediction model for POCD after TKR based on BN algorithm has good performance and high accuracy

关键词

Keywords

references

都义日, 米卫东. 术后认知功能障碍的研究进展[J]. 重庆医学, 2019, 48(6): 1021-1024. DOI: 10.3969/j.issn.1671-8348.2019.06.032.

DU Y R, MI W D. Research progress of postoperative cognitive dysfunction[J]. Chongqing Med, 2019, 48(6): 1021-1024. DOI: 10.3969/j.issn.1671-8348.2019.06.032.

边志民. 老年肿瘤患者合并术后认知功能障碍的影响因素及诊断防治策略[J]. 中国现代医学杂志, 2021, 31(13): 1-4. DOI: 10.3969/j.issn.1005-8982.2021.13.001.

BIANZ M. Advances in prevention and treatment of elderly cancer patients with postoperative cognitive dysfunction[J]. China J Mod Med, 2021, 31(13): 1-4. DOI: 10.3969/j.issn.1005-8982.2021.13.001.

TOPOL E J. High-performance medicine: the convergence of human and artificial intelligence[J]. Nat Med, 2019, 25(1): 44-56. DOI: 10.1038/s41591-018-0300-7.

RAVIZZA S, HUSCHTO T, ADAMOV A, et al . Predicting the early risk of chronic kidney disease in patients with diabetes using real-world data[J ] . Nat Med, 2019, 25(1): 57-59. DOI: 10.1038/s41591-018-0239-8.

KLANN J G, SZOLOVITS P, DOWNS S M, et al . Decision support from local data: creating adaptive order menus from past clinician behavior[J ] . J Biomed Inform, 2014, 48: 84-93. DOI: 10.1016/j.jbi.2013.12.005.

GRIMES D A, SCHULZ K F. An overview of clinical research: the lay of the land[J]. Lancet, 2002, 359(9300): 57-61. DOI: 10.1016/S0140-6736(02)07283-5.

支鸿羽. 基于机器学习的腹部手术患者术后死亡风险预测的研究[D]. 重庆: 中国人民解放军陆军军医大学, 2020.

ZHI H Y. Prediction of postoperative risk of death in patients undergoing abdominal surgery based on machine learning[D]. Chongqing : Army Medical University, 2020.

邓鹏, 陈芋文, 李雨捷, 等. 利用支持向量机预测重症监护室患者死亡风险的研究[J]. 陆军军医大学学报, 2022, 44(17): 1764-1769. DOI: 10.16016/j.2097-0927.202206112.

DENG P, CHEN Y W, LI Y J, et al . Prediction of in-hospital mortality risk in intensive care unit with support vector machine[J ] . J Army Med Univ, 2022, 44(17): 1764-1769. DOI: 10.16016/j.2097-0927.202206112.

ABBASS R, BHATTI U, ASINGER S. Bayesian networks: making the most of a history[J]. Clin Teach, 2021, 18(2): 140-141. DOI: 10.1111/tct.13263.

KYRIMI E, DUBE K, FENTON N, et al . Bayesian networks in healthcare: what is preventing their adoption?[J ] . Artif Intell Med, 2021, 116: 102079. DOI: 10.1016/j.artmed.2021.102079.

LUCAS P J F, VAN DER GAAG L C, ABU-HANNA A. Bayesian networks in biomedicine and health-care[J]. Artif Intell Med, 2004, 30(3): 201-214. DOI: 10.1016/j.artmed.2003.11.001.

SANDRI M, BERCHIALLA P, BALDI I, et al . Dynamic Bayesian Networks to predict sequences of organ failures in patients admitted to ICU[J ] . J Biomed Inform, 2014, 48: 106-113. DOI: 10.1016/j.jbi.2013.12.008.

PEELEN L, DE KEIZER N F, JONGE E. Using hierarchical dynamic Bayesian networks to investigate dynamics of organ failure in patients in the Intensive Care Unit[J]. J Biomed Inform, 2010, 43(2): 273-286. DOI: 10.1016/j.jbi.2009.10.002.

SESEN M B, NICHOLSON A E, BANARES-ALCANTARA R, et al . Bayesian networks for clinical decision support in lung cancer care[J ] . PLoS One, 2013, 8(12): e82349. DOI: 10.1371/journal.pone.0082349.

SESEN M B, PEAKE M D, BANARES-ALCANTARA R, et al . Lung Cancer Assistant: a hybrid clinical decision support application for lung cancer care[J ] . J R Soc Interface, 2014, 11(98): 20140534. DOI: 10.1098/rsif.2014.0534.

KUMAR M, VERMA G. Machine learning intrusion detection system based on MLP and RNN stochastic optimization technology[J]. J Res Sci Eng, 2022, 4(6): 6-9. DOI: 10.53469/jrse.2022.04(06).02.

ILIADIS L, NIKOLAIDIS S, SARIGIANNIDIS P, et al . Artwork style recognition using vision transformers and MLP mixer[J ] . Technologies, 2021, 10(1): 2. DOI: 10.3390/technologies10010002.

KLANN J, SCHADOW G, MCCOY J M. A recommendation algorithm for automating corollary order generation[J]. AMIA Annu Symp Proc, 2009, 2009: 333-337.

KLANN J, SCHADOW G, DOWNS S M. A method to compute treatment suggestions from local order entry data[J]. AMIA Annu Symp Proc, 2010, 2010: 387-391.

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Establishment of a chest-CT-image-based classification deep learning model for pneumonia prediction

Establishment of prediction model for risk of postoperative cognitive dysfunction after non-cardiac surgery based on different machine learning algorithms

Etiological factors for coronary heart disease and construction of a risk prediction model using two-sample MR and interpretable machine learning

Development and validation of a prediction model for amputation risk in patients with diabetic foot ulcers based on systematic review and meta-analysis

Predictive nomogram for central lymph node metastasis in papillary thyroid microcarcinoma based on CT-clinical data

Related Author

Meng ZENG

Na ZHAO

Xianqi WANG

Huilin PANG

Wenjing HOU

Lingling LIU

Xiaowen WANG

Jiali GONG

Related Institution

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

Chongqing Zhijian Life Technology Co., Ltd

Department of Radiology, Chongqing University Three Gorges Hospital

Department of Health Statistics， College of Military Preventive Medicine, University of International Business and Economics

Brigade of Non-Military Students， College of Basic Medical Sciences， Army Medical University （Third Military Medical University）， Chongqing, University of International Business and Economics

⁰