计算机科学 ›› 2024, Vol. 51 ›› Issue (6A): 230800091-7.doi: 10.11896/jsjkx.230800091
王彦麟1, 孙静1, 杨宏波2, 郭涛2, 潘家华2, 王威廉1
WANG Yanlin1, SUN Jing1, YANG Hongbo2, GUO Tao2, PAN Jiahua2, WANG Weilian1
摘要： 先心病相关肺动脉高压是一种严重的心血管疾病,致死率高,对其进行早期筛查与识别对于治愈尤为重要。目前临床是通过右心导管术确诊,此为有创检查,不便于在大规模筛查中采用,研究一种无创便捷的识别方法迫在眉睫。文中建立了一种时频融合的心音分类模型。首先对心音信号进行预处理,然后使用融合滤波器组对信号进行转换并求取动态时频特征,最后将得到的融合特征参数输入表格式先验数据拟合网络(TabPFN)中进行分类识别。实验结果表明,该算法在正常、CHD-PAH和CHD中的平均准确率、精确率、灵敏度、特异度和F1分别为92.21%,92.15%,92.15%,96.11%,92.14%。对于先心病相关肺动脉高压的早期筛查与识别具有重要意义。
中图分类号:
| [1]KEARNEY K,KOTLYAR E,LAU E M T.Pulmonary vascular disease as a systemic and multisystem disease[J].Clinics in Chest Medicine,2021,42(1):167-177. [2]NAEJIE R,RICHTER M J,RUBIN L J.The physiological basis of pulmonary arterial hypertension[J].European Respiratory Journal,2022,59(6):2334-2354. [3]GALIE N,HUMBERT M,VACHIERY J L,et al.2015 ESC/ERS guidelines for the diagnosis and treatment of pulmonary hypertension:the joint task force for the diagnosis and treatment of pulmonary hypertension of the European Society of Cardiology(ESC) and the European Respiratory Society(ERS):endorsed by:Association for European Paediatric and Congenital Cardio-logy(AEPC),International Society for Heart and Lung Transplantation(ISHLT)[J].European Heart Journal,2016,37(1):67-119. [4]YAGHI S,NOVIKOV A,TRANDAFIRESCU T.Clinical up-date on pulmonary hypertension[J].Journal of Investigative Medicine,2020,68(4):821-827. [5]MARON B A,KOVACS G,VAIDYA A,et al.Cardiopulmonary hemodynamics in pulmonary hypertension and heart failure:JACC review topic of the week[J].Journal of the American College of Cardiology,2020,76(22):2671-2681. [6]THOMAS C A,ANDERSON R J,CONDON D F,et al.Diagno-sis and management of pulmonary hypertension in the modern era:insights from the 6th world symposium[J].Pulmonary Therapy,2020,6(1):9-22. [7]SOMMER N,GHOFRANI H A,PAK O,et al.Current and future treatments of pulmonary arterial hypertension[J].British Journal of Pharmacology,2021,178(1):6-30. [8]PASCALL E,TULLOH R M R.Pulmonary hypertension incongenital heart disease[J].Future Cardiology,2018,14(4):343-353. [9]SIMONNEAU G,MONTANI D,CELERMAJER D S,et al.Haemodynamic definitions and updated clinical classification of pulmonary hypertension[J].European Respiratory Journal,2019,53(1):1913-1926. [10]SHUVO S B,ALAM SS,AYMAN S U,et al.NRC-Net:Automated noise robust cardio net for detecting valvular cardiac diseases using optimum transformation method with heart sound signals[J].Biomedical Signal Processing and Control,2023,86(1):105272. [11]TANG H,RONG Y,CHAI L,et al.Deep Learning Radar forHigh-Fidelity Heart Sound Recovery in Real-World Scenarios[J].IEEE Sensors Journal,2023,23:17803-17814. [12]WANG W B,QIN D M,WANG S B,et al.A multi-channelUNet framework based on SNMF-DCNN for robust heart-lung-sound separation[J].Computers in Biology and Medicine,2023,164:10. [13]ELOLA A,ARAMENDI E,OLIVEIRA J,et al.Beyond Heart Murmur Detection:Automatic Murmur Grading From Phonocardiogram[J].IEEE Journal of Biomedical and Health Informatics,2023,27:3856-3866. [14]GE B B,YANG H B,MA P Y,et al.Detection of pulmonary hypertension associated with congenital heart disease based on time-frequency domain and deep learning features[J].Biome-dical Signal Processing and Control,2023,81:11. [15]WANG X Z,YANG H B,ZONG R,et al.Heart sound classification based on sub-band envelope and convolution neural network[J].Journal of Biomedical Engineering,2021,38(5):969-978. [16]NIZAM N B,NUHASH S I S K,HASAN T.Hilbert-envelopefeatures for cardiac disease classification from noisy phonocardiograms[J].Biomedical Signal Processing and Control,2022,78(1):253-262. [17]KAMSON A P,SHARMA L N,DANDAPAT S.Multi-centroid diastolic duration distribution based HSMM for heart sound segmentation[J].Biomedical Signal Processing and Control,2019,48(1):265-272. [18]YILDIRIM M.Automatic classification and diagnosis of heartvalve diseases using heart sounds with MFCC and proposed deep model[J/OL].Concurrency and Computation:Practice and Experience(2022-07-25).https://onlinelibrary.wiley.com/doi/10.1002/cpe.7232. [19]CHEN P,ZHANG Q.Classification of heart sounds using discrete time-frequency energy feature based on S transform and the wavelet threshold denoising[J].Biomedical Signal Proces-sing and Control,2020,57(1):348-358. [20]DAR J A,SRIVASTAVA K K.,LONE S A.Spectral features and optimal Hierarchical attention networks for pulmonary abnormality detection from the respiratory sound signals[J].Biomedical Signal Processing and Control,2022,78(1):355-372. [21]KUI H,PAN J,ZONG R,et al.Heart sound classification based on log Mel-frequency spectral coefficients features and convolutional neural networks[J].Biomedical Signal Processing and Control,2021,69(1):175-190. [22]GUPTA C N,PALANIAPPAN R,SWAMINATHAN S,et al.Neural network classification of homomorphic segmented heart sounds[J].Applied Soft Computing,2007,7(1):286-297. [23]TANG H,JIANG Y,LI T,et al.Identification of pulmonary hypertension using entropy measure analysis of heart sound signal[J].Entropy,2018,20(5):389-405. [24]LI W H,WANG W B,GONG W G.Acoustic feature extraction of anomalous sounds in public spaces under low signal-to-noise ratio[J].Acta Acustica,2019,44(5):934-944. [25]GHAFFAR M S B A,KHAN U S,IQBAL J,et al.Improving classification performance of four class FNIRS-BCI using Mel Frequency Cepstral Coefficients(MFCC)[J].Infrared Physics & Technology,2021,112(1):425-436. [26]HOLLMANN N,MULLER S,EGGENSPERGER K,et al.TabPFN:A transformer that solves small tabular classification problems in a second[J/OL](2022-11-29).https://arxiv.org/abs/2207.01848. |
|
||
首页