基于神经网络多项生命特征的脓毒症预测方法
Sepsis Prediction Method Based on Neural Network Combined with Multiple Life Characteristics
DOI: 10.12677/MOS.2023.123212, PDF,   
作者: 何亚君, 潘 睿：上海理工大学机械工程学院，上海
关键词: 脓毒症；神经网络；CNN-LSTM；CLASS_WEIGHT；机器学习；Sepsis； Deep Neural Network； CNN-LSTM； CLASS_WEIGHT； Machine Learning
摘要: 针对缺乏ICU诊断条件的受伤患者，无法提前预测脓毒症感染风险的问题，提出了一种基于多项生命特征的神经网络(深度神经网络，CNN-LSTM)建立脓毒症预测模型的方法。为了解决数据严重不平衡的问题，引入了CLASS_WEIGHT技术。该模型使用医院ICU病患数据集进行训练和测试，与传统医学诊断得分制(例如SIRS、SOFA等)以及早期用于预测脓毒症的机器学习方法进行了比较。在AUROC评分指标上，该模型取得了较高的得分，表明该方法是有效和可行的。
Abstract: A method has been proposed to address the problem of being unable to predict sepsis infection risk in injured patients lacking ICU diagnostic conditions. The method involves building a sepsis predic-tion model based on multiple vital signs using a deep neural network (CNN-LSTM). The CLASS_WEIGHT technique was introduced to address severe data imbalance. The model was trained and tested using ICU patient datasets from hospitals, and was compared with traditional medical diagnosis scoring systems (such as SIRS and SOFA) and early machine learning methods for pre-dicting sepsis. The model achieved a high score on the AUROC evaluation metric, indicating that the approach is effective and feasible.
文章引用：何亚君, 潘睿. 基于神经网络多项生命特征的脓毒症预测方法[J]. 建模与仿真, 2023, 12(3): 2306-2317. https://doi.org/10.12677/MOS.2023.123212

参考文献

[1] 姚咏明, 盛志勇, 林洪远, 柴家科. 脓毒症定义及诊断的新认识[J]. 中国危重病急救医学, 2004(6): 321-324.
[2] Fleuren, L.M., Klausch, T.L.T., et al. (2020) Machine Learning for the Prediction of Sepsis: A Systematic Re-view and Meta-Analysis of Diagnostic Test Accuracy. Intensive Care Medicine, 46, 383-400. [Google Scholar] [CrossRef] [PubMed]
[3] Desautels, T., Calvert, J., et al. (2016) Prediction of Sepsis in the Inten-sive Care Unit with Minimal Electronic Health Record Data: A Machine Learning Approach. JMIR Medical Informatics, 4, e28. [Google Scholar] [CrossRef] [PubMed]
[4] Seymour, C.W., Liu, V.X., Iwashyna, T.J., et al. (2016) Assessment of Clinical Criteria for Sepsis: For the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA, 315, 762-774. [Google Scholar] [CrossRef] [PubMed]
[5] Vincent, J.L. (2008) Clinical Sepsis and Septic Shock—Definition, Diagno-sis and Management Principles. Langenbeck’s Archives of Surgery, 393, 817-824. [Google Scholar] [CrossRef] [PubMed]
[6] Johnson, A.E., Pollard, T.J., Shen, L., et al. (2016) MIMIC-III, a Freely Accessible Critical Care Database. Scientific Data, 3, Article ID: 160035. [Google Scholar] [CrossRef] [PubMed]
[7] Singer, M., Deutschmanc, C.S., Seymourc, C.W., et al. (2016) The Third In-ternational Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA, 315, 801-810. [Google Scholar] [CrossRef] [PubMed]
[8] Shankar-Hari, M., Phillips, G.S., Levy, M.L., et al. (2016) Developing a New Definition and Assessing New Clinical Criteria for Septic Shock: For the Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA, 315, 775-787. [Google Scholar] [CrossRef] [PubMed]
[9] Perner, A., Cec-coni, M., Cronhjort, M., et al. (2018) Expert Statement for the Management of Hypovolemiain Sepsis. Intensive Care Medicine, 44, 791-798. [Google Scholar] [CrossRef] [PubMed]
[10] Prescott, H.C. and Angus, D.C. (2018) Post Sepsis Mor-bidity. JAMA, 319, 91. [Google Scholar] [CrossRef] [PubMed]
[11] Reinhart, K., Daniels, R., Kissoon, N., et al. (2017) Recognizing Sepsis as a Global Health Priority—A WHO Resolution. The New England Journal of Medicine, 377, 414-417. [Google Scholar] [CrossRef] 
[12] Shi, X.J., Chen, Z.R., et al. (2015) A Machine Learning Approach for Pre-cipitation Nowcasting.
[13] Royal College of Physicians (2012) National Early Warning Score (NEWS): Standardising the As-sessment of Acute Illness Severity in the NHS. Report of a Working Party, RCP, London.
[14] Nitish, S., Geoffrey, E., et al. (2014) A Simple Way to Prevent Neural Networks from Overfitting. Journal of Machine Learning Research, 15, 1929-1958.
[15] Longadge, R. and Dongre, S. (2013) Class Imbalance Problem in Data Mining Review.
[16] Raith, E.P., Udy, A.A., Bailey, M., et al. (2017) Prognostic Accuracy of the SOFA Score, SIRS Criteria, and qSOFA Score for In-Hospital Mortality among Adults with Suspected Infection Admitted to the Intensive Care Unit. JAMA, 317, 290-300. [Google Scholar] [CrossRef] [PubMed]
[17] Fawcett, T. (2006) An Introduction to ROC Analysis. Pattern Recognition Letters, 27, 861-874. [Google Scholar] [CrossRef] 
[18] Kam, H.J. and Kim, H.Y. (2017) Learning Representations for the Early Detection of Sepsis with Deep Neural Networks. Computers in Biology and Medicine, 89, 248-255. [Google Scholar] [CrossRef] [PubMed]
[19] Calvert, J.S., Price, D.A., et al. (2016) A Computational Ap-proach to Early Sepsis Detection. Computers in Biology and Medicine, 74, 69-73. [Google Scholar] [CrossRef] [PubMed]
[20] Faisal, M., Scally, A., Richardson, D., et al. (2018) Development and External Validation of an Automated Computer-Aided Risk Score for Predicting Sepsis in Emergency Medical Admissions Using the Patient’s First Electronically Recorded Vital Signs and Blood Test Results. Critical Care Medicine, 46, 612-618. [Google Scholar] [CrossRef] 
[21] Horng, S., Sontag, D.A., Halpern, Y., et al. (2017) Creating an Automated Trigger for Sepsis Clinical Decision Support at Emergency Department Triage Using Machine Learning. PLOS ONE, 12, e0174708. [Google Scholar] [CrossRef] [PubMed]
[22] Mao, Q., Jay, M., Hoffman, J.L., et al. (2018) Multicentre Validation of a Sepsis Prediction Algorithm Using Only Vital Sign Data in the Emergency Department, General Ward and ICU. BMJ Open, 8, e017833. [Google Scholar] [CrossRef] [PubMed]
[23] Nemati, S., Holder, A., Razmi, F., et al. (2018) An Interpretable Ma-chine Learning Model for Accurate Prediction of Sepsis in the ICU. Critical Care Medicine, 46, 547-553. [Google Scholar] [CrossRef] 

为你推荐