Predicting fetal hypoxia using common spatial pattern and machine learning from cardiotocography signals

Abstract Cardiotocography (CTG) is a screening tool used in daily obstetric practice to determine fetal wellbeing. Its interpretation is generally performed visually by the field experts, and this visual inspection is an error-prone and subjective process. In addition, it leads to several drawbacks, such as variability among the observers and low reproducibility rates. To tackle these drawbacks, a novel computer-aided diagnostic (CAD) model is proposed. As novel diagnostic indices, the features provided by the common spatial patterns (CSP) were considered in this study. The experiments were carried out on a publicly available CTU-UHB Intrapartum CTG database. Four different data division criteria were evaluated individually. The proposed model relied upon a combination of the conventional as well as the CSP features and machine learning models such as an artificial neural network (ANN), support vector machine (SVM), and k -nearest neighbor ( k NN). To validate the successes of the models, the five-fold cross-validation method was employed. The results validated that the CSP features ensured an increase in the performances of the machine learning models in the fetal hypoxia detection task. Also, the most effective results were provided by the SVM classifier with an accuracy of 94.75%, a sensitivity of 74.29% and a specificity of 99.55%. Consequently, thanks to the proposed model, a novel, consistent, and robust diagnostic model ensured for predicting fetal hypoxia.

[1]  Yang Zhang,et al.  Computer-Aided Diagnosis System of Fetal Hypoxia Incorporating Recurrence Plot With Convolutional Neural Network , 2019, Front. Physiol..

[2]  Lee-Wen Huang,et al.  Spontaneous fetal heart rate deceleration: an ominous sign for fetal outcome , 2002, Archives of Gynecology and Obstetrics.

[3]  Corinna Cortes,et al.  Support-Vector Networks , 1995, Machine Learning.

[4]  G. Pfurtscheller,et al.  Optimal spatial filtering of single trial EEG during imagined hand movement. , 2000, IEEE transactions on rehabilitation engineering : a publication of the IEEE Engineering in Medicine and Biology Society.

[5]  Ersen Yilmaz,et al.  Determination of Fetal State from Cardiotocogram Using LS-SVM with Particle Swarm Optimization and Binary Decision Tree , 2013, Comput. Math. Methods Medicine.

[6]  Zafer Cömert,et al.  Comparison of Machine Learning Techniques for Fetal Heart Rate Classification , 2017 .

[7]  J. M. Swartjes,et al.  Computer analysis of antepartum fetal heart rate: 2. Detection of accelerations and decelerations. , 1990, International journal of bio-medical computing.

[8]  Agnese Sbrollini,et al.  Extraction of Digital Cardiotocographic Signals from Digital Cardiotocographic Images: Robustness of eCTG Procedure , 2019, Electronics.

[9]  Umit Budak,et al.  A Simple and Effective Approach for Digitization of the CTG Signals from CTG Traces , 2019, IRBM.

[10]  Lenka Lhotská,et al.  Open access intrapartum CTG database , 2014, BMC Pregnancy and Childbirth.

[11]  Maria Romano,et al.  Analysis of Foetal Heart Rate Variability Components by Means of Empirical Mode Decomposition , 2016 .

[12]  Agnieszka Wosiak,et al.  Supervised and unsupervised machine learning for improved identification of intrauterine growth restriction types , 2016, 2016 Federated Conference on Computer Science and Information Systems (FedCSIS).

[13]  P. Warrick,et al.  Neural network based detection of fetal heart rate patterns , 2005, Proceedings. 2005 IEEE International Joint Conference on Neural Networks, 2005..

[14]  Edmond Zahedi,et al.  Antepartum fetal heart rate feature extraction and classification using empirical mode decomposition and support vector machine , 2011, Biomedical engineering online.

[15]  G. Moody,et al.  Power spectral density of unevenly sampled data by least-square analysis: performance and application to heart rate signals , 1998, IEEE Transactions on Biomedical Engineering.

[16]  Sandro Fioretti,et al.  eCTG: an automatic procedure to extract digital cardiotocographic signals from digital images , 2018, Comput. Methods Programs Biomed..

[17]  Zafer Cömert,et al.  Open-access software for analysis of fetal heart rate signals , 2018, Biomed. Signal Process. Control..

[18]  D. Ayres-de-Campos,et al.  FIGO consensus guidelines on intrapartum fetal monitoring: Cardiotocography , 2015, International journal of gynaecology and obstetrics: the official organ of the International Federation of Gynaecology and Obstetrics.

[19]  Lenka Lhotská,et al.  Using nonlinear features for fetal heart rate classification , 2012, Biomed. Signal Process. Control..

[20]  Carlo Cattani,et al.  On the Creation of a New Diagnostic Model for Fetal Well-Being on the Base of Wavelet Analysis of Cardiotocograms , 2006, Journal of Medical Systems.

[21]  Maria Romano,et al.  Prognostic Decision Support Using Symbolic Dynamics in CTG Monitoring , 2013, EFMI-STC.

[22]  Petar M. Djuric,et al.  Fetal Heart Rate Classification Using Generative Models , 2014, IEEE Transactions on Biomedical Engineering.

[23]  M. C. Terek,et al.  Different types of variable decelerations and their effects to neonatal outcome. , 2003, Singapore medical journal.

[24]  Lenka Lhotská,et al.  Investigating pH based evaluation of fetal heart rate (FHR) recordings , 2017, Health and Technology.

[25]  H W Jongsma,et al.  Analysis of heart rate and beat-to-beat variability: Interval difference index. , 1980, American journal of obstetrics and gynecology.

[26]  Zafer Cömert,et al.  Prediction of intrapartum fetal hypoxia considering feature selection algorithms and machine learning models , 2019, Health Information Science and Systems.

[27]  J. Álvarez-Ramírez,et al.  Fractal and nonlinear changes in the long-term baseline fluctuations of fetal heart rate. , 2012, Medical engineering & physics.

[28]  Berat Dogan Optimized spatial filters as a new method for mass spectrometry-based cancer diagnosis , 2016, Appl. Soft Comput..

[29]  Tareq Manzoor,et al.  A Strategy for Classification of “Vaginal vs. Cesarean Section” Delivery: Bivariate Empirical Mode Decomposition of Cardiotocographic Recordings , 2019, Front. Physiol..

[30]  Francesco Amato,et al.  Evaluation of floatingline and foetal heart rate variability , 2018, Biomed. Signal Process. Control..

[31]  George A. Macones,et al.  Intrapartum fetal heart rate monitoring: nomenclature, interpretation, and general management principles , 2009 .

[32]  Ana Paula Rocha,et al.  Linear and nonlinear fetal heart rate analysis of normal and acidemic fetuses in the minutes preceding delivery , 2006, Medical and Biological Engineering and Computing.

[33]  Sepideh Hajipour Sardouie,et al.  Extended common spatial and temporal pattern (ECSTP): A semi-blind approach to extract features in ERP detection , 2019, Pattern Recognit..

[34]  Zafer Cömert,et al.  Evaluation of Fetal Distress Diagnosis during Delivery Stages based on Linear and Nonlinear Features of Fetal Heart Rate for Neural Network Community , 2016 .

[35]  Patrice Abry,et al.  Information Theory to Probe Intrapartum Fetal Heart Rate Dynamics , 2017, Entropy.

[36]  Riccardo Bellazzi,et al.  Comparison of data mining techniques applied to fetal heart rate parameters for the early identification of IUGR fetuses , 2016, 2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC).

[37]  Abdulhamit Subasi,et al.  Classification of the cardiotocogram data for anticipation of fetal risks using machine learning techniques , 2015, Appl. Soft Comput..

[38]  Kemal Polat,et al.  Automatic determination of digital modulation types with different noises using Convolutional Neural Network based on time-frequency information , 2020, Appl. Soft Comput..

[39]  Austin Ugwumadu,et al.  Agreement and accuracy using the FIGO, ACOG and NICE cardiotocography interpretation guidelines , 2017, Acta obstetricia et gynecologica Scandinavica.

[40]  Luis Filipe Coelho Antunes,et al.  Entropy and Compression Capture Different Complexity Features: The Case of Fetal Heart Rate , 2017, Entropy.

[41]  Kemal Polat,et al.  Binary particle swarm optimization (BPSO) based channel selection in the EEG signals and its application to speller systems , 2020, Journal of Artificial Intelligence and Systems.

[42]  Yanhui Guo,et al.  Classification of multi-carrier digital modulation signals using NCM clustering based feature-weighting method , 2019, Comput. Ind..

[43]  Zhongheng Zhang,et al.  Introduction to machine learning: k-nearest neighbors. , 2016, Annals of translational medicine.

[44]  Umit Budak,et al.  Efficient approach for digitization of the cardiotocography signals , 2020 .

[45]  Ana Paula Rocha,et al.  Linear and nonlinear analysis of heart rate patterns associated with fetal behavioral states in the antepartum period. , 2007, Early human development.

[46]  Dimitrios Kassanos,et al.  Wavelet analysis and neural networks for intrapartum fetal monitoring. Can we long for a new technique? Is it doable? , 2008, Medical science monitor : international medical journal of experimental and clinical research.

[47]  Mei-Ling Huang,et al.  Fetal distress prediction using discriminant analysis, decision tree, and artificial neural network , 2012 .

[48]  Cömert Zafer,et al.  Fusing fine-tuned deep features for recognizing different tympanic membranes , 2020 .

[49]  Yue Ruan,et al.  Quantum Algorithm for K-Nearest Neighbors Classification Based on the Metric of Hamming Distance , 2017, International Journal of Theoretical Physics.

[50]  Philip A. Warrick,et al.  Antenatal fetal heart rate acceleration detection , 2016, 2016 Computing in Cardiology Conference (CinC).

[51]  A. Cahill,et al.  Electronic fetal monitoring: past, present, and future. , 2011, Clinics in perinatology.

[52]  Burhan Ergen,et al.  A Deep Feature Learning Model for Pneumonia Detection Applying a Combination of mRMR Feature Selection and Machine Learning Models , 2020, IRBM.

[53]  Xiaoping Li,et al.  Common Spatio-Temporal Pattern for Single-Trial Detection of Event-Related Potential in Rapid Serial Visual Presentation Triage , 2011, IEEE Transactions on Biomedical Engineering.

[54]  Engin Avci,et al.  A Diagnostic Model for Identification of Myocardial Infarction from Electrocardiography Signals , 2017 .

[55]  Amparo Alonso-Betanzos,et al.  The NST-EXPERT project: the need to evolve , 1995, Artif. Intell. Medicine.

[56]  Zafer Cömert,et al.  Detection of lung cancer on chest CT images using minimum redundancy maximum relevance feature selection method with convolutional neural networks , 2020 .

[57]  Kemal Polat,et al.  A novel demodulation structure for quadrate modulation signals using the segmentary neural network modelling , 2020 .

[58]  Yefei Zhang,et al.  DeepFHR: intelligent prediction of fetal Acidemia using fetal heart rate signals based on convolutional neural network , 2019, BMC Medical Informatics and Decision Making.

[59]  Nihat Daldal,et al.  A novel demodulation method for quadrate type modulations using a hybrid signal processing method , 2020 .

[60]  J. De Jonckheere,et al.  Inter-observer reliability of 4 fetal heart rate classifications. , 2017, Journal of gynecology obstetrics and human reproduction.

[61]  Zafer Cömert,et al.  Waste classification using AutoEncoder network with integrated feature selection method in convolutional neural network models , 2020 .

[62]  H. Geijn 2 Developments in CTG analysis , 1996 .

[63]  Maria Romano,et al.  Foetal heart rate variability frequency characteristics with respect to uterine contractions , 2010 .

[64]  E. Chandraharan,et al.  Continuous cardiotocography during labour: Analysis, classification and management. , 2016, Best practice & research. Clinical obstetrics & gynaecology.

[65]  Jeffrey M. Hausdorff,et al.  Physionet: Components of a New Research Resource for Complex Physiologic Signals". Circu-lation Vol , 2000 .

[66]  R. Gonzalez-Camarena,et al.  Analysis of high frequency fetal heart rate variability using empirical mode decomposition , 2005, Computers in Cardiology, 2005.

[67]  G S Dawes,et al.  System 8000: Computerized antenatal FHR analysis , 1991, Journal of perinatal medicine.

[68]  Zafer Cömert,et al.  Using wavelet transform for cardiotocography signals classification , 2017, 2017 25th Signal Processing and Communications Applications Conference (SIU).

[69]  Zafer Cömert,et al.  Prognostic model based on image-based time-frequency features and genetic algorithm for fetal hypoxia assessment , 2018, Comput. Biol. Medicine.

[70]  J. P. Marques de Sá,et al.  The Porto system for automated cardiotocographic signal analysis , 1991, Journal of perinatal medicine.

[71]  Kemal Polat,et al.  Deep Learning Applications for Hyperspectral Imaging: A Systematic Review , 2020, Journal of the Institute of Electronics and Computer.