| 106 | 0 | 25 |
| 下载次数 | 被引频次 | 阅读次数 |
为减轻长期动态心电图判读的负担并降低误诊率，提出了一种基于超短时心率变异性（ultra-short-term heart rate variability, UST-HRV）的心律失常分类方法。通过MIT-BIH数据库系统提取了26项时域、频域及非线性HRV特征，评估30 s UST-HRV与300 s HRV的一致性，并构建了基于LightGBM的融合时间平铺、不平衡处理和个体差异校正的自动心律失常分类模型。结果表明：20项UST-HRV特征对于300 s HRV具有良好替代性；通过改进交叉验证与时间平铺策略，模型实现了正常窦性、房早、室早、房颤和房扑的自动识别，性能达到准确率99.09%、精确率99.53%、召回率99.52%、F1分数99.44%，说明UST-HRV可利用短序列数据表征自主神经调控功能，实现良好的心律失常辨识，为快速筛查和减轻长期动态心电图判读负担提供智能化方案。
Abstract：To reduce the burden of long-term Holter interpretation and improve arrhythmia screening, an arrhythmia classification method based on ultra-short-term heart rate variability(UST-HRV) was proposed. From the MIT-BIH database, 26 time-domain, frequency-domain, and nonlinear HRV features were extracted to assess consistency between 30 s UST-HRV and 300 s HRV. An automatic arrhythmia classification model based on LightGBM was developed with time tiling, imbalance handling, and individual difference correction. Results show that 20 USTHRV features are effective substitutes for 300 s HRV. With improved cross-validation and time tiling, the model automatically identifies normal sinus rhythm, premature atrial contractions, premature ventricular contractions, atrial fibrillation, and atrial flutter, achieving an accuracy of 99.09%, a precision of 99.53%, a recall of 99.52%, and an F1-score of 99.44%. The results demonstrate that UST-HRV can effectively characterize autonomic nervous regulation with short-sequence data, enable effective arrhythmia identification, and provide an intelligent solution for rapid screening and reducing Holter interpretation burden.
[1]Kaptoge S, Pennells L, De Bacquer D, et al. World Health Organization cardiovascular disease risk charts:revised models to estimate risk in 21 global regions[J]. The Lancet Global Health, 2019, 7(10):e1332-e1345.
[2]陈璇,王雨锋,张筑欣,等.中国心律失常现状及治疗进展[J].中国研究型医院, 2020, 7(1):75-78, 198-201.Chen Xuan, Wang Yufeng, Zhang Zhuxin, et al. Current status and management of arrhythmia in China[J]. Journal of Chinese Research Hospitals, 2020, 7(1):75-78, 198-201(in Chinese).
[3]Saraiya A, Corsi D, Farhan S, et al. Abstract 4142981:incidence and predictors of symptom-arrhythmia association during holter monitoring[J]. Circulation, 2024, 150(Suppl_1):4142981.
[4]Joyce J J, Bogarapu S, Odhiambo C, et al. Holter monitor rhythm parameters in healthy infants, children, and adolescents:defining reference limits with meta-analysis[J]. Journal of the American Heart Association, 2025, 14(11):e039783.
[5]Do Nascimento P C G, Martins M S, Improta-Caria A C, et al.Applicability of machine learning algorithms in diagnosis of atrial fibrillation and LQTS by electrocardiogram interpretation:a systematic review[J]. Arquivos Brasileiros de Cardiologia, 2025, 122(8):e20240843.
[6]Guan C J, Gong A W, Zhao Y, et al. Interpretable machine learning model for new-onset atrial fibrillation prediction in critically ill patients:a multi-center study[J]. Critical Care,2024, 28:349.
[7]Sharma A, Dhanka S, Kumar A, et al. A comparative study of heterogeneous machine learning algorithms for arrhythmia classification using feature selection technique and multidimensional datasets[J]. Engineering Research Express, 2024,6(3):035209.
[8]Kolk M Z H, Deb B, Ruipérez-Campillo S, et al. Machine learning of electrophysiological signals for the prediction of ventricular arrhythmias:systematic review and examination of heterogeneity between studies[J]. eBioMedicine, 2023, 89:104462.
[9]Jeong D U, Taye G T, Hwang H J, et al. Optimal length of heart rate variability data and forecasting time for ventricular fibrillation prediction using machine learning[J]. Computational and Mathematical Methods in Medicine, 2021, 2021:6663996.
[10]Simon S T, Mandair D, Tiwari P, et al. Prediction of druginduced long QT syndrome using machine learning applied to harmonized electronic health record data[J]. Journal of Cardiovascular Pharmacology and Therapeutics, 2021, 26(4):335-340.
[11]Deng J W, Ma J R, Yang J, et al. Design and implementation of ECG classification based on heart rate variability analysis[C]//2024 IEEE 4th International Conference on Information Technology, Big Data and Artificial Intelligence(ICIBA). December 6-8, 2024, Chongqing,China. IEEE, 2024:1594-1599.
[12]Drenjancevic I, Grizelj I, Harsanji-Drenjancevic I, et al.The interplay between sympathetic overactivity, hypertension and heart rate variability(Review, invited)[J].Acta Physiologica Hungarica, 2014, 101(2):129-142.
[13]Malik M, Bigger J T, Camm A J, et al. Heart rate variability:Standards of measurement, physiological interpretation, and clinical use[J]. European Heart Journal,1996, 17(3):354-381.
[14]Thong T, Li K, McNames J, et al. Accuracy of ultrashort heart rate variability measures[C]//Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Cancun,Mexico:IEEE, 2003:2424-2427.[LinkOut]
[15]Munoz M L, van Roon A, Riese H, et al. Validity of(ultra-)short recordings for heart rate variability measurements[J]. PLoS One, 2015, 10(9):e0138921.
[16]Wu L, Shi P, Yu H L, et al. An optimization study of the ultra-short period for HRV analysis at rest and postexercise[J]. Journal of Electrocardiology, 2020, 63:57-63.
[17]Burma J S, Graver S, Miutz L N, et al. The validity and reliability of ultra-short-term heart rate variability parameters and the influence of physiological covariates[J]. Journal of Applied Physiology, 2021, 130(6):1848-1867.
[18]Ebrahimzadeh E, Foroutan A, Shams M, et al. An optimal strategy for prediction of sudden cardiac death through a pioneering feature-selection approach from HRV signal[J]. Computer Methods and Programs in Biomedicine, 2019, 169:19-36.
[19]Wehler D, Jelinek H F, Gronau A, et al. Reliability of heartrate-variability features derived from ultra-short ECG recordings and their validity in the assessment of cardiac autonomic neuropathy[J]. Biomedical Signal Processing and Control,2021, 68:102651.
[20]Chieng T M, Hau Y W, Omar Z, et al. Validity and reliability of the ultra-short-term heart rate variability features in predicting ventricular tachyarrhythmia[J]. Biomedical Signal Processing and Control, 2025, 110:108173.
[21]Moody G B, Mark R G. The impact of the MIT-BIH arrhythmia database[J]. IEEE Engineering in Medicine and Biology Magazine, 2001, 20(3):45-50.
[22]Moody G B. A new method for detecting atrial fibrillation using R-R intervals[J]. Computers in Cardiology, 1983, 10:227-230.
[23]Goldberger A L, Amaral L A N, Glass L, et al. PhysioBank,PhysioToolkit, and PhysioNet:components of a new research resource for complex physiologic signals[J]. Circulation,2000, 101(23):e215-e220.
[24]Scargle J D. Studies in astronomical time series analysis. IIStatistical aspects of spectral analysis of unevenly spaced data[J]. The Astrophysical Journal, 1982, 263:835-853.
[25]Fonseca D S, Netto A D, Ferreira R B, et al. Lomb-scargle periodogram applied to heart rate variability study[C]//2013ISSNIP Biosignals and Biorobotics Conference:Biosignals and Robotics for Better and Safer Living(BRC). Rio de Janeiro, Brazil:IEEE, 2013:1-4.
[26]Ke Guolin, Meng Qi, Finley T, et al. LightGBM:A highly efficient gradient boosting decision tree[J]. Neural Information Processing Systems, 2017, 30:3815895.
基本信息:
中图分类号:TP181;TN911.7;R541.7
引用信息:
[1]赵喆,唐建同,王天昊,等.基于超短时HRV特征的心律失常识别模型及有效性验证[J].天津工业大学学报().
基金信息:
国家重点研发研发计划项目（2023YFC3011802）
2026-04-17
2026-04-17
2026-04-17