|
基于CNN网络的孕妇腹电信号质量评估算法
|
Abstract:
胎儿心率(fetal heart rate, FHR)能够反映母亲子宫内胎儿的健康情况,也是胎儿监护的重要指标。无创胎儿心电监测是将电极置于孕妇腹部来采集腹部心电信号(abdomen electrocardiogram, AECG),经过信号处理可从AECG获取出胎儿心电信号(fetal electrocardiogram, FECG),根据获得的FECG可计算得到FHR。目前FECG提取存在着一些问题,如在AECG有母亲心电信号(maternal electrocardiogram, MECG)、基线漂移、工频干扰、采集噪声等,这些噪声会使AECG的信号质量差,最终影响FECG的提取效果。因此本文提出了一种基于CNN网络的AECG质量评估算法,通过该算法可以对AECG质量评估,筛选出信号质量比较好的AECG,从而提高FECG提取的准确率。通过对测试集的AECG质量评估,本文提出算法的灵敏度(SE)、阳性预测值(PPV)和F1值达到了97.76%、97.00%、97.38%,证明本文提出的方法可有效地对AECG质量评估。
Fetal heart rate (FHR) can reflect the health of the foetus in the mother’s womb and is an important indicator of foetal monitoring. Non-invasive fetal electrocardiography is based on placing electrodes on the abdomen of a pregnant woman to collect abdominal electrocardiogram (AECG), and after signal processing, fetal electrocardiogram (FECG) can be obtained from the AECG signals, and FHR can be calculated based on the FECG signals obtained. Currently there are some problems in FECG separation, such as maternal electrocardiogram (MECG) in AECG, baseline drift, IF interference, acquisition noise, etc., which will make the signal quality of AECG poor and ultimately affect the separation effect of FECG. Therefore, this paper proposes an AECG quality assessment algorithm based on deep learning CNN network, by which the AECG quality can be assessed and the AECGs with better signal quality can be screened out, so as to improve the accuracy of FECG extraction. By evaluating the AECG quality of the test set, the sensitivity (SE), positive predictive value (PPV), and F1 value of the proposed algorithm in this paper reached 97.76%, 97.00%, and 97.38%, which proves that the method proposed in this paper can be effective for AECG quality evaluation.
[1] | Martinek, R., Kahankova, R., Jezewski, J., et al. (2018) Comparative Effectiveness of ICA and PCA in Extraction of Fetal ECG from Adominal Signals: Toward Non-Invasive Fetal Monitoring. Frontiers in Physiology, 9, Article 648. https://doi.org/10.3389/fphys.2018.00648 |
[2] | Yu, Q., Yan, H., Song, L., et al. (2018) Automatic Identifying of Maternal ECG Source When Applying ICA in Fetal ECG Extraction. Biocybernetics and Biomedical Engineering, 38, 448-455. https://doi.org/10.1016/j.bbe.2018.03.003 |
[3] | 彭朗鸣, 郭家松, 尹春艳. 诊断超声对胎儿的影响[J]. 中国医疗前沿, 2007, 2(12): 68-69. |
[4] | Goldberger, A.L., Amaral, L., Glass, L., et al. (2000) PhysioBank, PhysioToolkit, and PhysioNet—Components of a New Research Resource for Complex Physiologic Signals. Circulation, 101, E215-E220. https://doi.org/10.1161/01.CIR.101.23.e215 |
[5] | Podziemski, P. and Gieraltowski, J. (2013) Fetal Heart Rate Discovery: Algorithm for Detection of Fetal Heart Rate from Noisy, Noninvasive Fetal ECG Recordings. Computing in Cardiology 2013, Zaragoza, 22-25 September 2013, 333-336. |
[6] | Graatsma, E.M. (2010) Monitoring of Fetal Heart Rate and Uterine Activity. University Utrecht, Utrecht. |
[7] | Silva, I., Behar, J., Sameni, R., et al. (2013) Noninvasive Fetal ECG: The PhysioNet/Computing in Cardiology Challenge 2013. Computing in Cardiology 2013, Zaragoza, 22-25 September 2013, 149-152. |
[8] | Martinek, R., Kahankova, R., Nazeran, H., et al. (2017) Non-Invasive Fetal Monitoring: A Maternal Surface ECG Electrode Place Ment-Based Novel Approach for Optimization of Adaptive Filter Control Parameters Using the LMS and RLS Algorithms. Sensors, 17, Article 1154. https://doi.org/10.3390/s17051154 |
[9] | Grzywalski, T. and Drgas, S. (2022) Speech Enhancement Using U-Nets with Wide-Context Units. Multimedia Tools and Applications, 81, 18617-18639. https://doi.org/10.1007/s11042-022-12632-6 |
[10] | Kang, Z., Huang, Z. and Lu, C. (2022) Speech Enhancement Using U-Net with Compressed Sensing. Applied Sciences, 12, Article 4161. https://doi.org/10.3390/app12094161 |
[11] | Lee, K.J. and Lee, B. (2022) End-to-End Deep Learning Architecture for Separating Maternal and Fetal ECGs Using W-Net. IEEE Access, 10, 39782-39788. https://doi.org/10.1109/ACCESS.2022.3166925 |
[12] | Wang, X., Han, Y. and Deng, Y. (2023) CSGSA-Net: Canonical-Structured Graph Sparse Attention Network for Fetal ECG Estimation. Biomedical Signal Processing and Control, 82, Article 104556. https://doi.org/10.1016/j.bspc.2022.104556 |
[13] | Zhang, J.-M., Huang, X.-L., Guan, Q., Liu, T.-B., Li, P., et al. (2015) Some Regularity on How to Locate Electrodes for Higher fECG SNRs. Chinese Physics B, 24, 442-446. https://doi.org/10.1088/1674-1056/24/3/038702 |
[14] | Andreotti, F., Grer, F., Malberg, H., et al. (2017) Non-Invasive Fetal ECG Signal Quality Assessment for Multichannel Heart Rate Estimation. IEEE Transactions on Biomedical Engineering, 64, 2793-2802. https://doi.org/10.1109/TBME.2017.2675543 |
[15] | Nardelli, M., Lanata, A., Valenza, G., et al. (2020) A Tool for the Real-Time Evaluation of ECG Signal Quality and Activity: Application to Submaximal Treadmill Test in Horses. Biomedical Signal Processing and Control, 56, Article 101666. https://doi.org/10.1016/j.bspc.2019.101666 |
[16] | Shi, Y., Han, N., Li, P., et al. (2019) Robust Assessment of ECG Signal Quality for Wearable Devices. 2019 IEEE International Conference on Healthcare Informatics (ICHI), Xi’an, 10-13 June 2019, 1-3. https://doi.org/10.1109/ICHI.2019.8904810 |
[17] | Andreotti, F., Behar, J., Zaunseder, S., et al. (2016) An Open-Source Framework for Stress-Testing Non-Invasive Foetal ECG Extraction Algorithms. Physiological measurement, 37, Article 627. https://doi.org/10.1088/0967-3334/37/5/627 |
[18] | Behar, J., Andreotti, F., Zaunseder, S., et al. (2016) A Practical Guide to Non-Invasive Foetal Electrocardiogram Extraction and Analysis. Physiological Measurement, 37, R1-R35. https://doi.org/10.1088/0967-3334/37/5/R1 |
[19] | 郭垚垚, 陈兆学. 一种脉搏波和心电信号时域基线漂移消除方法[J]. 中国医学物理学杂志, 2016, 33(2): 167-172. |
[20] | Fotiadou, E., Konopczyński, T., Hesser, J.W., et al. (2020) End-to-End Trained CNN Encoder-Decoder Network for Fetal ECG Signal Denoising. Physiological Measurement, 41, Article 015005. https://doi.org/10.1088/1361-6579/ab69b9 |
[21] | Ye, W., Jiang, Z., Li, Q., et al. (2022) A Hybrid Model for Pathological Voice Recognition of Post-Stroke Dysarthria by Using 1DCNN and Double-LSTM Networks. Applied Acoustics, 197, Article 108934. https://doi.org/10.1016/j.apacoust.2022.108934 |
[22] | Liusong, Y. and Hui, D. (2022) Voice Quality Evaluation of Singing Art Based on 1DCNN Model. Mathematical Problems in Engineering, 2022, Article ID: 2074844. https://doi.org/10.1155/2022/2074844 |
[23] | Huang, S., Tang, J., Dai, J., et al. (2019) Signal Status Recognition Based on 1DCNN and Its Feature Extraction Mechanism Analysis. Sensors, 19, Article 2018. https://doi.org/10.3390/s19092018 |
[24] | 刘伟, 单雪垠, 李双喜, 等. 基于并行1DCNN的滚动轴承故障诊断研究[J]. 机电工程, 2021, 38(12): 1572-1578. |
[25] | 肖斌, 陈嘉博, 毕秀丽, 等. 基于一维卷积神经网络与循环神经网络串联的心音分析方法[J]. 电子学报, 2022, 50(10): 2425-2432. |