Fetal ECG extraction from maternal ECG using deeply supervised LinkNet plus plus model

被引:10
|
作者
Rahman, Arafat [1 ]
Mahmud, Sakib [2 ]
Chowdhury, Muhammad E. H. [2 ]
Yalcin, Huseyin Cagatay [3 ]
Khandakar, Amith [2 ]
Mutlu, Onur [3 ]
Bin Mahbub, Zaid [4 ]
Kamal, Reema Yousef [5 ]
Pedersen, Shona [6 ]
机构
[1] Univ Dhaka, Dept Biomed Phys & Technol, Dhaka 1000, Bangladesh
[2] Qatar Univ, Dept Elect Engn, Doha 2713, Qatar
[3] Qatar Univ, Biomed Res Ctr, Doha, Qatar
[4] North South Univ, Dept Phys & Math, Dhaka, Bangladesh
[5] Hamad Med Corp, Hamad Gen Hosp, Pediat Cardiol Div, Doha, Qatar
[6] Qatar Univ, Coll Med, Dept Basic Med Sci, Doha 2713, Qatar
关键词
Fetal ECG (fECG); Mother ECG (mECG); Coronary Heart Disease (CHD); Signal reconstruction; Deep supervision; Deep Learning; Convolutional Neural Network (CNN); HEART-RATE; SIGNALS;
D O I
10.1016/j.engappai.2023.106414
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Fetal heart monitoring and early disease detection using non-invasive fetal electrocardiograms (fECG) can help substantially to reduce infant death through improved diagnosis of Coronary Heart Disease (CHD) in the fetus. Despite the potential benefits, non-invasive fECG extraction from maternal abdominal ECG (mECG) is a challenging problem due to multiple factors such as the overlap of maternal and fetal R-peaks, low amplitude of fECG, and various systematic and environmental noises. Conventional fECG extraction techniques, such as adaptive filters, independent component analysis (ICA), empirical mode decomposition (EMD), etc., face various performance issues due to the fECG extraction challenges. In this paper, we proposed a novel deep learning architecture, LinkNet++ (motivated by the original LinkNet) to extract fECG from abdominal mECG automatically and efficiently using two different publicly available datasets. LinkNet++ is equipped with a feature-addition method to combine deep and shallow levels with residual blocks to overcome the limitations of U-Net and UNet++ models. It also has deep supervised and densely connected convolution blocks to overcome the limitations of the original LinkNet. The proposed LinkNet++ model was evaluated using fECG signal reconstruction and fetal QRS (fQRS) detection. As a signal-to-signal synthesis model, LinkNet++ performed very well in two real-life datasets and achieved 85.58% and 87.60% Pearson correlation coefficients (PCC) between the ground truth and predicted fECG on two datasets, respectively. In terms of fQRS detection, it also outperformed most of the previous works and showed excellent performance with more than 99% of F1 scores on both datasets. Our results indicate that the proposed model can potentially extract fECG non-invasively with excellent signal quality, thereby providing an excellent diagnostic tool for various fetal heart diseases.
引用
收藏
页数:14
相关论文
共 50 条
  • [1] Extraction of Fetal ECG from Maternal ECG
    Serdengecti, Cigdem
    Engin, Mehmet
    Engin, Erkan Zeki
    Balci, Soner
    BIYOMUT: 2009 14TH NATIONAL BIOMEDICAL ENGINEERING MEETING, 2009, : 416 - 418
  • [2] Fetal ECG Extraction From Maternal ECG Using Attention-Based CycleGAN
    Mohebbian, Mohammad Reza
    Vedaei, Seyed Shahim
    Wahid, Khan A.
    Dinh, Anh
    Marateb, Hamid Reza
    Tavakolian, Kouhyar
    IEEE JOURNAL OF BIOMEDICAL AND HEALTH INFORMATICS, 2022, 26 (02) : 515 - 526
  • [3] Fetal ECG extraction from single-channel maternal ECG using singular value decomposition
    Kanjilal, PP
    Palit, S
    Saha, G
    IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 1997, 44 (01) : 51 - 59
  • [4] A Supervised ICA Algorithm for Fetal ECG Extraction
    Tsao, Jenho
    Hsu, Chao-Yang
    Lo, Men-Tzung
    ISSNIP 2008: PROCEEDINGS OF THE 2008 INTERNATIONAL CONFERENCE ON INTELLIGENT SENSORS, SENSOR NETWORKS, AND INFORMATION PROCESSING, 2008, : 327 - +
  • [5] The Maternal Abdominal ECG as Input to MICA in the Fetal ECG Extraction Problem
    Camargo-Olivares, J. L.
    Martin-Clemente, R.
    Hornillo-Mellado, S.
    Elena, M. M.
    Roman, I.
    IEEE SIGNAL PROCESSING LETTERS, 2011, 18 (03) : 161 - 164
  • [6] nn The maternal ECG suppression algorithm for efficient extraction of the fetal ECG from abdominal signal
    Matonia, A.
    Jezewski, J.
    Horoba, K.
    Gacek, A.
    Labaj, P.
    2006 28TH ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE AND BIOLOGY SOCIETY, VOLS 1-15, 2006, : 296 - 299
  • [7] Non-invasive Fetal ECG Extraction from Maternal Abdominal ECG Using LMS and RLS Adaptive Algorithms
    Kahankova, Radana
    Martinek, Radek
    Bilik, Petr
    PROCEEDINGS OF THE THIRD INTERNATIONAL AFRO-EUROPEAN CONFERENCE FOR INDUSTRIAL ADVANCEMENT-AECIA 2016, 2018, 565 : 258 - 271
  • [8] A novel fetal ecg signal extraction from maternal ecg signal using conditional generative adversarial networks (CGAN)
    Vadivu, M. Senthil
    Kavithaa, G.
    JOURNAL OF INTELLIGENT & FUZZY SYSTEMS, 2022, 43 (01) : 801 - 811
  • [9] Principal component model for maternal ECG extraction in fetal QRS detection
    Lipponen, Jukka A.
    Tarvainen, Mika P.
    PHYSIOLOGICAL MEASUREMENT, 2014, 35 (08) : 1637 - 1648
  • [10] Suppression of maternal ECG from fetal ECG using neuro fuzzy logic technique
    Vijila, CKS
    Renganathan, S
    Johnson, S
    PROCEEDINGS OF THE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS 2003, VOLS 1-4, 2003, : 1007 - 1012