计算机科学

计算机科学 ›› 2025, Vol. 52 ›› Issue (9): 106-118.doi: 10.11896/jsjkx.250300037

• 智能医学工程 • 上一篇    下一篇

基于机器学习的介入式葡萄糖传感器故障监测模型

刘思行1,2, 许硕洋3, 徐鹤1,3, 季一木1,3   

  1. 1 江苏省高性能计算与智能处理工程研究中心 南京 210023
    2 江苏鱼跃医疗设备股份有限公司 江苏 镇江 212300
    3 南京邮电大学计算机学院、软件学院、网络空间安全学院 南京 210023
  • 收稿日期:2025-03-10 修回日期:2025-04-24 出版日期:2025-09-15 发布日期:2025-09-11
  • 通讯作者: 徐鹤(xuhe@njupt.edu.cn)
  • 作者简介:(liu.sx@yuyue.com.cn)
  • 基金资助:
    江苏鱼跃医疗设备股份有限公司科技项目(2022外017);江苏省研究生实践创新计划项目(SJCX23_0274)

Machine Learning Based Interventional Glucose Sensor Fault Monitoring Model

LIU Sixing1,2, XU Shuoyang3, XU He1,3, JI Yimu1,3   

  1. 1 Jiangsu HPC and Intelligent Processing Engineer Research Center,Nanjing 210023,China
    2 Jiangsu Yuyue Medical Equipment and Supply Co.,Ltd.,Zhenjiang,Jiangsu 212300,China
    3 School of Computer Science,Nanjing University of Posts and Telecommunications,Nanjing 210023,China
  • Received:2025-03-10 Revised:2025-04-24 Online:2025-09-15 Published:2025-09-11
  • About author:LIU Sixing,born in 1991,postgraduate.His main research interests include artificial intelligence,bioelectrical signals,biochemical signal analysis and proces-sing.
    XU He,born in 1985,Ph.D,professor,master supervisor,is a senior member of CCF(No.19957S).His main research interests include artificial intelligence and big data,Internet of Things technology.
  • Supported by:
    Science and Technology Project of Jiangsu Yuyue Medical Equipment Co.,Ltd.(2022外017) and Jiangsu Provincial Graduate Practice and Innovation Program Project(SJCX23_0274).

PDF (PC)

11

摘要： 随着传感器技术的进步,血糖监测从传统的单点采集发展为连续动态监测(CGM),通过介入式葡萄糖传感器实时监测间质液葡萄糖浓度。血糖传感器的运行状态对监测准确性至关重要,但传感器故障识别面临类别不平衡问题,导致机器学习模型性能下降。基于此,提出了一种结合数据预处理、特征工程和模型集成的优化策略。首先,通过缺失值填补和噪声处理提升数据的完整性和可靠性;其次,利用合成少数类过采样技术(SMOTE)对少数类样本进行过采样,缓解类别不平衡问题;最后,采用堆叠泛化(Stacking)的集成学习方法,结合基于焦点损失函数(Focal Loss)优化的极端梯度提升(XGBoost)和类别特征梯度提升(CatBoost)集成基分类器,与逻辑回归(LR)元分类器构建双层模型,进一步提升故障监测的准确性。为了证明所提出模型的有效性,将该模型的预测结果与其他模型进行了对比,包括基于Focal Loss的单一XGBoost,及其分别与SVM,KNN,LightGBM作为基分类器构建的集成模型等。研究结果表明,提出的基于Focal Loss的XGBoost和 CatBoost模型在传感器故障分类任务中表现良好,PR曲线和ROC曲线效果均优于其他模型,精确度和召回率分别为0.925 0和0.923 8。

关键词: 传感器故障监测, 堆叠泛化, 集成学习, 极端梯度提升, 类别特征梯度提升

Abstract: With the advancement of sensor technology,blood glucose monitoring has evolved from traditional single-point sampling to continuous dynamic monitoring CGM,enabling real-time monitoring of interstitial fluid glucose concentration through interventional glucose sensor.The operational status of glucose sensors is crucial for monitoring accuracy,but sensor fault detection faces the challenge of class imbalance,leading to degraded performance of machine learning models.Based on this,this paper proposes an optimization strategy that combines data preprocessing,feature engineering,and model integration.Firstly,the completeness and reliability of the data are improved through missing value imputation and noise reduction.Secondly,the SMOTE is used to oversample minority class samples,alleviating the class imbalance problem.Finally,a two-layer model is constructed using the Stacking ensemble learning method,which combines base classifiers of XGBoost optimized with Focal Loss and CatBoost with a LR meta-classifier,further enhancing the accuracy of fault monitoring.To demonstrate the effectiveness of the proposed model,its prediction results are compared with other models,including a single XGBoost model based on Focal Loss and ensemble models constructed with SVM,KNN,and LightGBM as base classifiers.The results show that the proposed XGBoost optimized with the Focal Loss and CatBoost models,perform well in the sensor fault classification task,with both PR and ROC curves outperforming other models,achieving precision and recall rates of 0.925 0 and 0.923 8,respectively.

Key words: Sensor fault detection, Stacked generalization, Ensemble learning, Extreme Gradient Boosting, Categorical Boosting

中图分类号: 

  • TP391
[1]TEHRANI F,TEYMOURIAN H,WUERSTLE B,et al.An Integrated Wearable Microneedle Array for the Continuous Monitoring of Multiple Biomarkers in Interstitial Fluid[J].Nature Biomedical Engineering,2022,6(11):1214-1224.
[2]WANG X T,LIU X Y.Fault Detection for Electro Hydraulic Position Servo System Based on Kernel Regression Interval Model[J].Journal of Xinyang Normal University(Natural Science Edition),2024,37(2):252-258.
[3]MAAN H,ZHANG L,YU C,et al.Characterizing the Impacts of Dataset Imbalance on Single-Cell Data Integration[J].Nature Biotechnology,2024,42:1899-1908.
[4]LE T T H,OKTIAN Y E,KIM H.XGBoost for ImbalancedMulticlass Classification-Based Industrial Internet of Things Intrusion Detection Systems[J].Sustainability,2022,14(14):8707.
[5]FAN C,LI C,PENG Y,et al.Fault Diagnosis of Vibration Sensors Based on Triage Loss Function-Improved XGBoost[J].Electronics,2023,12(21):4442.
[6]WANG C,DENG C,WANG S.Imbalance-XGBoost:LeveragingWeighted and Focal Losses for Binary Label-Imbalanced Classification with XGBoost[J].Pattern Recognition Letters,2020,136:190-197.
[7]WANG Z,LOU S,LIANG S,et al.Multi-Class DisturbanceEvents Recognition Based on EMD and XGBoost in Φ-Otdr[J].IEEE Access,2020,8:63551-63558.
[8]BEN JABEUR S,STEF N,CARMONA P.Bankruptcy Prediction Using the XGBoost Algorithm and Variable Importance Feature Engineering[J].Computational Economics,2023,61(2):715-741.
[9]SHARMA H,HARSORA H,OGUNLEYE B.An OptimalHouse Price Prediction Algorithm:XGBoost[J].Analytics,2024,3(1):30-45.
[10]OUKHOUYA H,KADIRI H,EL HIMDI K,et al.Forecasting International Stock Market Trends:XGBoost,LSTM,LSTM-XGBoost,and Backtesting XGBoost Models[J].Statistics,Optimization & Information Computing,2024,12(1):200-209.
[11]ZHAO C,PENG R,WU D.Bagging and Boosting Fine-Tuning for Ensemble Learning[J].IEEE Transactions on Artificial Intelligence,2023,5(4):1728-1742.
[12]MONDAL S,GHOSH S,NAG A.Brain Stroke Prediction Mo-del Based on Boosting and Stacking Ensemble Approach[J].International Journal of Information Technology,2024,16(1):437-446.
[13]WONG W Y,HASIKIN K,KHAIRUDDIN M,et al.A Stacked Ensemble Deep Learning Approach for Imbalanced Multi-Class Water Quality Index Prediction[J].Computers Materials and Continua,2023,76:1361-1384.
[14]SHUVO M M H,ISLAM S K.Deep Multitask Learning by Stacked Long Short-Term Memory for Predicting Personalized Blood Glucose Concentration[J].IEEE Journal of Biomedical and Health Informatics,2023,27(3):1612-1623.
[15]AL-GHATTAS O,BAO J,SANZ-ALONSO D.Ensemble Kalman Filters with Resampling[J].SIAM/ASA Journal on Uncertainty Quantification,2024,12(2):411-441.
[16]LANCE G,LEROY T,SERY J.Adaptive Extended Kalman Filter for Pemfc Membrane Water Content Estimation[J].International Journal of Hydrogen Energy,2024,71:1164-1173.
[17]WONGVORACHAN T,HE S,BULUT O.A Comparison ofUndersampling,Oversampling,and SMOTE Methods for Dealing with Imbalanced Classification in Educational Data Mining[J].Information,2023,14(1):54.
[18]KHLEEL N A A,NEHÉZ K.A Novel Approach for Software Defect Prediction Using CNN and GRU Based on SMOTE Tomek Method[J].Journal of Intelligent Information Systems,2023,60(3):673-707.
[19]ELREEDY D,ATIYA A F,KAMALOV F.A Theoretical Distribution Analysis of Synthetic Minority Oversampling Technique(SMOTE) for Imbalanced Learning[J].Machine Lear-ning,2024,113(7):4903-4923.
[20]KISHOR A,CHAKRABORTY C.Early and Accurate Prediction of Diabetics Based on Fcbf Feature Selection and SMOTE[J].International Journal of System Assurance Engineering and Management,2024,15(10):4649-4657.
[21]BANO A,KÜNZLER J,WEHRLI F,et al.Clinical Evidence for High-Risk Ce-Marked Medical Devices for Glucose Management:A Systematic Review and Meta-Analysis[J].Diabetes,Obesity and Metabolism,2024,26(10):4753-4766.
[22]YI M K,LEE W K,HWANG S O.A Human Activity Recognition Method Based on Lightweight Feature Extraction Combined with Pruned and Quantized CNN for Wearable Device[J].IEEE Transactions on Consumer Electronics,2023,69(3):657-670.
[23]BUCKLEY T,GHOSH B,PAKRASHI V.A Feature Extraction & Selection Benchmark for Structural Health Monitoring[J].Structural Health Monitoring,2023,22(3):2082-2127.
[24]AL-HADDAD L A,JABER A A,HAMZAH M N,et al.Vibration-Current Data Fusion and Gradient Boosting Classifier for Enhanced Stator Fault Diagnosis in Three-Phase Permanent Magnet Synchronous Motors[J].Electrical Engineering,2024,106(3):3253-3268.
[25]WANG Y.Fractional Fourier Transform and Its Application[J].Theoretical and Natural Science,2024,42:8-12.
[26]ZHANG Z,LI Y,WANG G,et al.Supervised Mineral Prospectivity Mapping Via Class-Balanced Focal Loss Function on Imbalanced Geoscience Datasets[J].Mathematical Geosciences,2023,55(7):989-1010.
[27]HOSSAIN M S,BETTS J M,PAPLINSKI A P.Dual Focal Loss to Address Class Imbalance in Semantic Segmentation[J].Neurocomputing,2021,462:69-87.
[28]TIAN J,TSAI P W,ZHANG K,et al.Synergetic Focal Loss for Imbalanced Classification in Federated XGBoost[J].IEEE Transactions on Artificial Intelligence,2023,5(2):647-660.
[29]PASUPA K,VATATHANAVARO S,TUNGJITNOB S.Con-volutional Neural Networks Based Focal Loss for Class Imba-lance Problem:A Case Study of Canine Red Blood Cells Morphology Classification[J].Journal of Ambient Intelligence and Humanized Computing,2023,14(11):15259-15275.
[30]EL-GHANDOUR M,OBAYYA M I.Pneumonia Detection inChest X-Ray Images Using an Optimized Ensemble with XGBoost Classifier[J].Multimedia Tools and Applications,2024,84:5491-5521.
[31]DING H,YANG X Q,QI T Y.Hybrid Machine Learning Mo-dels Based on CatBoost Classifier for Assessing Students’ Academic Performance[J].International Journal of Advanced Computer Science and Applications,2024,15(7):94-106.
[32]SUN Y.Explainable Prediction of Compressive Strength and Elastic Modulus for Concrete Containing Waste Foundry Sand Using Bayesian-Optimized XGBoost with 10-Fold Cross-Validation[J].Journal of Sustainable Metallurgy,2024,10(1):335-359.
[33]ULLAH I,LIU K,YAMAMOTO T,et al.Electric Vehicle Energy Consumption Prediction Using Stacked Generalization:An Ensemble Learning Approach[J].International Journal of Green Energy,2021,18(9):896-909.
[34]QARAEI M,BABBAR R.Meta-Classifier Free Negative Sam-pling for Extreme Multilabel Classification[J].Machine Lear-ning,2024,113(2):675-697.
[35]AVCI D,SERT E,DOGANTEKIN E,et al.A New Super Resolution Faster R-CNN Model Based Detection and Classification of Urine Sediments[J].Biocybernetics and Biomedical Enginee-ring,2023,43(1):58-68.
[36]CHANG W,WANG X,YANG J,et al.An Improved CatBoost-Based Classification Model for Ecological Suitability of Blueberries[J].Sensors,2023,23(4):1811.
[37]KURANI A,DOSHI P,VAKHARIA A,et al.A Comprehensive Comparative Study of Artificial Neural Network(ANN) and Support Vector Machines(SVM) on Stock Forecasting[J].Annals of Data Science,2023,10(1):183-208.
[38]MOHY-EDDINE M,GUEZZAZ A,BENKIRANE S,et al.An Efficient Network Intrusion Detection Model for Iot Security Using KNN Classifier and Feature Selection[J].Multimedia Tools and Applications,2023,82(15):23615-23633.
[39]FANG Q,SHEN B,XUE J.A New Elite Opposite SparrowSearch Algorithm-Based Optimized LightGBM Approach for Fault Diagnosis[J].Journal of Ambient Intelligence and Humanized Computing,2023,14(8):10473-10491.
[40]LIN L,MA X,CHEN C,et al.Imbalanced Industrial Load Identification Based on Optimized CatBoost with Entropy Features[J].Journal of Electrical Engineering & Technology,2024,19:4817-4832.
[41]ZHUANG H,LEHNER F,DE GAETANO A T.Improved Diagnosis of Precipitation Type with LightGBM Machine Learning[J].Journal of Applied Meteorology and Climatology,2024,63(3):437-453.
[42]BAHRANI P,MINAEI-BIDGOLI B,PARVIN H,et al.A New Improved KNN-Based Recommender System[J].The Journal of Supercomputing,2024,80(1):800-834.
[43]ASSELMAN A,KHALDI M,AAMMOU S.Enhancing the Prediction of Student Performance Based on the Machine Learning XGBoost Algorithm[J].Interactive Learning Environments,2023,31(6):3360-3379.
[44]LIANG H Y.Fault Diagnosis of Power Transformer Based on Stacked Sparse Autoencoder and XGBoost[J].Journal of Chongqing Technology and Business University(Natural Science Edition),2024,41(6):65-71.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市渝北区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发