A novel algorithm based on ensemble empirical mode decomposition for non-invasive fetal ECG extraction.

Non-invasive fetal electrocardiography appears to be one of the most promising fetal monitoring techniques during pregnancy and delivery nowadays. This method is based on recording electrical potentials produced by the fetal heart from the surface of the maternal abdomen. Unfortunately, in addition to the useful fetal electrocardiographic signal, there are other interference signals in the abdominal recording that need to be filtered. The biggest challenge in designing filtration methods is the suppression of the maternal electrocardiographic signal. This study focuses on the extraction of fetal electrocardiographic signal from abdominal recordings using a combination of independent component analysis, recursive least squares, and ensemble empirical mode decomposition. The method was tested on two databases, the Fetal Electrocardiograms, Direct and Abdominal with Reference Heartbeats Annotations and the PhysioNet Challenge 2013 database. The evaluation was performed by the assessment of the accuracy of fetal QRS complexes detection and the quality of fetal heart rate determination. The effectiveness of the method was measured by means of the statistical parameters as accuracy, sensitivity, positive predictive value, and F1-score. Using the proposed method, when testing on the Fetal Electrocardiograms, Direct and Abdominal with Reference Heartbeats Annotations database, accuracy higher than 80% was achieved for 11 out of 12 recordings with an average value of accuracy 92.75% [95% confidence interval: 91.19-93.88%], sensitivity 95.09% [95% confidence interval: 93.68-96.03%], positive predictive value 96.36% [95% confidence interval: 95.05-97.17%] and F1-score 95.69% [95% confidence interval: 94.83-96.35%]. When testing on the Physionet Challenge 2013 database, accuracy higher than 80% was achieved for 17 out of 25 recordings with an average value of accuracy 78.24% [95% confidence interval: 73.44-81.85%], sensitivity 81.79% [95% confidence interval: 76.59-85.43%], positive predictive value 87.16% [95% confidence interval: 81.95-90.35%] and F1-score 84.08% [95% confidence interval: 80.75-86.64%]. Moreover, the non-invasive ST segment analysis was carried out on the records from the Fetal Electrocardiograms, Direct and Abdominal with Reference Heartbeats Annotations database and achieved high accuracy in 7 from in total of 12 records (mean values µ < 0.1 and values of ±1.96σ < 0.1).

New Method for Beat-to-Beat Fetal Heart Rate Measurement Using Doppler Ultrasound Signal.

The most commonly used method of fetal monitoring is based on heart activity analysis. Computer-aided fetal monitoring system enables extraction of clinically important information hidden for visual interpretation-the instantaneous fetal heart rate (FHR) variability. Today's fetal monitors are based on monitoring of mechanical activity of the fetal heart by means of Doppler ultrasound technique. The FHR is determined using autocorrelation methods, and thus it has a form of evenly spaced-every 250 ms-instantaneous measurements, where some of which are incorrect or duplicate. The parameters describing a beat-to-beat FHR variability calculated from such a signal show significant errors. The aim of our research was to develop new analysis methods that will both improve an accuracy of the FHR determination and provide FHR representation as time series of events. The study was carried out on simultaneously recorded (during labor) Doppler ultrasound signal and the reference direct fetal electrocardiogram Two subranges of Doppler bandwidths were separated to describe heart wall movements and valve motions. After reduction of signal complexity by determining the Doppler ultrasound envelope, the signal was analyzed to determine the FHR. The autocorrelation method supported by a trapezoidal prediction function was used. In the final stage, two different methods were developed to provide signal representation as time series of events: the first using correction of duplicate measurements and the second based on segmentation of instantaneous periodicity measurements. Thus, it ensured the mean heart interval measurement error of only 1.35 ms. In a case of beat-to-beat variability assessment the errors ranged from -1.9% to -10.1%. Comparing the obtained values to other published results clearly confirms that the new methods provides a higher accuracy of an interval measurement and a better reliability of the FHR variability estimation.

Fetal electrocardiograms, direct and abdominal with reference heartbeat annotations.

Monitoring fetal heart rate (FHR) variability plays a fundamental role in fetal state assessment. Reliable FHR signal can be obtained from an invasive direct fetal electrocardiogram (FECG), but this is limited to labour. Alternative abdominal (indirect) FECG signals can be recorded during pregnancy and labour. Quality, however, is much lower and the maternal heart and uterine contractions provide sources of interference. Here, we present ten twenty-minute pregnancy signals and 12 five-minute labour signals. Abdominal FECG and reference direct FECG were recorded simultaneously during labour. Reference pregnancy signal data came from an automated detector and were corrected by clinical experts. The resulting dataset exhibits a large variety of interferences and clinically significant FHR patterns. We thus provide the scientific community with access to bioelectrical fetal heart activity signals that may enable the development of new methods for FECG signals analysis, and may ultimately advance the use and accuracy of abdominal electrocardiography methods.

Detection of Atrial Fibrillation Episodes in Long-Term Heart Rhythm Signals Using a Support Vector Machine.

Atrial fibrillation (AF) is a serious heart arrhythmia leading to a significant increase of the risk for occurrence of ischemic stroke. Clinically, the AF episode is recognized in an electrocardiogram. However, detection of asymptomatic AF, which requires a long-term monitoring, is more efficient when based on irregularity of beat-to-beat intervals estimated by the heart rate (HR) features. Automated classification of heartbeats into AF and non-AF by means of the Lagrangian Support Vector Machine has been proposed. The classifier input vector consisted of sixteen features, including four coefficients very sensitive to beat-to-beat heart changes, taken from the fetal heart rate analysis in perinatal medicine. Effectiveness of the proposed classifier has been verified on the MIT-BIH Atrial Fibrillation Database. Designing of the LSVM classifier using very large number of feature vectors requires extreme computational efforts. Therefore, an original approach has been proposed to determine a training set of the smallest possible size that still would guarantee a high quality of AF detection. It enables to obtain satisfactory results using only 1.39% of all heartbeats as the training data. Post-processing stage based on aggregation of classified heartbeats into AF episodes has been applied to provide more reliable information on patient risk. Results obtained during the testing phase showed the sensitivity of 98.94%, positive predictive value of 98.39%, and classification accuracy of 98.86%.

A Review of Signal Processing Techniques for Non-Invasive Fetal Electrocardiography.

Fetal electrocardiography (fECG) is a promising alternative to cardiotocography continuous fetal monitoring. Robust extraction of the fetal signal from the abdominal mixture of maternal and fetal electrocardiograms presents the greatest challenge to effective fECG monitoring. This is mainly due to the low amplitude of the fetal versus maternal electrocardiogram and to the non-stationarity of the recorded signals. In this review, we highlight key developments in advanced signal processing algorithms for non-invasive fECG extraction and the available open access resources (databases and source code). In particular, we highlight the advantages and limitations of these algorithms as well as key parameters that must be set to ensure their optimal performance. Improving or combining the current or developing new advanced signal processing methods may enable morphological analysis of the fetal electrocardiogram, which today is only possible using the invasive scalp electrocardiography method.

Is Abdominal Fetal Electrocardiography an Alternative to Doppler Ultrasound for FHR Variability Evaluation?

Great expectations are connected with application of indirect fetal electrocardiography (FECG), especially for home telemonitoring of pregnancy. Evaluation of fetal heart rate (FHR) variability, when determined from FECG, uses the same criteria as for FHR signal acquired classically-through ultrasound Doppler method (US). Therefore, the equivalence of those two methods has to be confirmed, both in terms of recognizing classical FHR patterns: baseline, accelerations/decelerations (A/D), long-term variability (LTV), as well as evaluating the FHR variability with beat-to-beat accuracy-short-term variability (STV). The research material consisted of recordings collected from 60 patients in physiological and complicated pregnancy. The FHR signals of at least 30 min duration were acquired dually, using two systems for fetal and maternal monitoring, based on US and FECG methods. Recordings were retrospectively divided into normal (41) and abnormal (19) fetal outcome. The complex process of data synchronization and validation was performed. Obtained low level of the signal loss (4.5% for US and 1.8% for FECG method) enabled to perform both direct comparison of FHR signals, as well as indirect one-by using clinically relevant parameters. Direct comparison showed that there is no measurement bias between the acquisition methods, whereas the mean absolute difference, important for both visual and computer-aided signal analysis, was equal to 1.2 bpm. Such low differences do not affect the visual assessment of the FHR signal. However, in the indirect comparison the inconsistencies of several percent were noted. This mainly affects the acceleration (7.8%) and particularly deceleration (54%) patterns. In the signals acquired using the electrocardiography the obtained STV and LTV indices have shown significant overestimation by 10 and 50% respectively. It also turned out, that ability of clinical parameters to distinguish between normal and abnormal groups do not depend on the acquisition method. The obtained results prove that the abdominal FECG, considered as an alternative to the ultrasound approach, does not change the interpretation of the FHR signal, which was confirmed during both visual assessment and automated analysis.

Application of fuzzy inference systems for classification of fetal heart rate tracings in relation to neonatal outcome.

OBJECTIVES: Fetal monitoring based on the analysis of the fetal heart rate (FHR) signal is the most common method of biophysical assessment of fetal condition during pregnancy and labor Visual analysis of FHR signals presents a challenge due to a complex shape of the waveforms. Therefore, computer-aided fetal monitoring systems provide a number of parameters that are the result of the quantitative analysis of the registered signals. These parameters are the basis for a qualitative assessment of the fetal condition. The guidelines for the interpretation of FHR provided by FIGO are commonly used in clinical practice. On their basis a weighted fuzzy scoring system was constructed to assess the FHR tracings using the same criteria as those applied by expert clinicians. The effectiveness of the automated classification was evaluated in relation to the fetal outcome assessed by Apgar score. MATERIAL AND METHODS: The proposed automated system for fuzzy classification is an extension of the scoring systems used for qualitative evaluation of the FHR tracings. A single fuzzy rule of the system corresponds to a single evaluation principle of a signal parameter derived from the FIGO guidelines. The inputs of the fuzzy system are the values of quantitative parameters of the FHR signal, whereas the system output, which is calculated in the process of fuzzy inference, defines the interpretation of the FHR tracing. The fuzzy evaluation process is a kind of diagnostic test, giving a negative or a positive result that can be compared with the fetal outcome assessment. The present retrospective study included a set of 2124 one-hour antenatal FHR tracings derived from 333 patients, recorded between 24 and 44 weeks of gestation (mean gestational age: 36 weeks). Various approaches for the research data analysis, depending on the method of interpretation of the individual patient-tracing relation, were used in the investigation. The quality of the fuzzy analysis was defined by the number of correct classifications (CC) and the additional index QI - the geometric mean of the sensitivity and specificity values. RESULTS: The effectiveness of the fetal assessment varied, depending on the assumed relation between a patient and a set of her tracings. The approach, based on a common assessment of the whole set of tracings recorded for a single patient, provided the highest quality of automated classification. The best results (CC = 70.9% and QI = 84.0%) confirmed the possibility of predicting the neonatal outcome using the proposed fuzzy system based on the FIGO guidelines. CONCLUSIONS: It is possible to enhance the process of the fetal condition assessment with classification of the FHR records through the implementation of the heuristic rules of inference in the fuzzy signal processing algorithms.

Predicting the risk of low-fetal birth weight from cardiotocographic signals using ANBLIR system with deterministic annealing and epsilon-insensitive learning.

Cardiotocography (CTG) is a biophysical method of fetal condition assessment based mainly on recording and automated analysis of fetal heart activity. The computerized fetal monitoring systems provide the quantitative description of the CTG signals, but the effective conclusion generation methods for decision process support are still needed. Assessment of the fetal state can be verified only after delivery using the fetal (newborn) outcome data. One of the most important features defining the abnormal fetal outcome is low birth weight. This paper describes an application of the artificial neural network based on logical interpretation of fuzzy if-then rules neurofuzzy system to evaluate the risk of low-fetal birth weight using the quantitative description of CTG signals. We applied different learning procedures integrating least squares method, deterministic annealing (DA) algorithm, and epsilon-insensitive learning, as well as various methods of input dataset modification. The performance was evaluated with the number of correctly classified cases (CC) expressed as the percentage of the testing set size, and with overall index (OI) being the function of predictive indexes. The best classification efficiency (CC = 97.5% and OI = 82.7%), was achieved for integrated DA with epsilon-insensitive learning and dataset comprising of the CTG traces recorded as earliest for a given patient. The obtained results confirm efficiency for supporting the fetal outcome prediction using the proposed methods.

Some practical remarks on neural networks approach to fetal cardiotocograms classification.

Cardiotocographic monitoring is a primary biophysical method for assessment of a fetal state based on quantitative analysis of the biophysical signals. Although the computerized fetal monitoring systems have become a standard in clinical centres, the effective methods, which could enable conclusion generation, are still being searched. In the proposed work the attempts have been made to answer some important questions, which occurred during application of neural network for classification of the fetal state as being normal or abnormal. These questions are particularly important for medical applications and concern the influence of data set organization, inputs representation and the network's architecture. The networks of MLP and RBF types were developed and tested using 50 trials, with randomly mixed data contents in learning, validating and testing subsets. Additionally, the influence of numerical and categorical representation of the input quantitative parameters describing fetal cardiotocograms on the efficiency of the learning process was tested.

New approach to quantitative description of deceleration of fetal heart rate for the patterns classification.

The most common method of biophysical fetal monitoring is recording and analyzing the cardiotocographic signals. In analysis of the fetal heart rate signal special emphasis is paid to the deceleration patterns and their correlation to the uterine contractions. According to deceleration classification the most important is the distinguishing between the periodic and the episodic types. In visual analysis, this classification is based on fuzzy description of deceleration onset being "abrupt" or "gradual". Application of commonly used interpretation of these imprecise terms in computer aided monitoring systems very often leads to erroneous classifications. Therefore, the redefinition of the deceleration nadir phase, as a group of samples around the lowest point, is required. It ensures that the onset phase, which is very important in deceleration classification, will consist of only appropriate samples. For determination of nadir the new method based on three stage-analysis of samples frequency distribution was developed. To evaluate the proposed method we compared the results with reference data obtained from clinical experts.