Evaluation of convolutional neural network for recognizing uterine contractions with electrohysterogram

Automatic semantic segmentation of EHG recordings by deep learning: An approach to a screening tool for use in clinical practice

BACKGROUND AND OBJECTIVE

Preterm delivery is an important factor in the disease burden of the newborn and infants worldwide. Electrohysterography (EHG) has become a promising technique for predicting this condition, thanks to its high degree of sensitivity. Despite the technological progress made in predicting preterm labor, its use in clinical practice is still limited, one of the main barriers being the lack of tools for automatic signal processing without expert supervision, i.e. automatic screening of motion and respiratory artifacts in EHG records. Our main objective was thus to design and validate an automatic system of segmenting and screening the physiological segments of uterine origin in EHG records for robust characterization of uterine myoelectric activity, predicting preterm labor and help to promote the transferability of the EHG technique to clinical practice.

METHODS

For this, we combined 300 EHG recordings from the TPEHG DS database and 69 EHG recordings from our own database (Ci2B-La Fe) of women with singleton gestations. This dataset was used to train and evaluate U-Net, U-Net++, and U-Net 3+ for semantic segmentation of the physiological and artifacted segments of EHG signals. The model's predictions were then fine-tuned by post-processing.

RESULTS

U-Net 3+ outperformed the other models, achieving an area under the ROC curve of 91.4 % and an average precision of 96.4 % in detecting physiological activity. Thresholds from 0.6 to 0.8 achieved precision from 93.7 % to 97.4 % and specificity from 81.7 % to 94.5 %, detecting high-quality physiological segments while maintaining a trade-off between recall and specificity. Post-processing improved the model's adaptability by fine-tuning both the physiological and corrupted segments, ensuring accurate artifact detection while maintaining physiological segment integrity in EHG signals.

CONCLUSIONS

As automatic segmentation proved to be as effective as double-blind manual segmentation in predicting preterm labor, this automatic segmentation tool fills a crucial gap in the existing preterm delivery prediction system workflow by eliminating the need for double-blind segmentation by experts and facilitates the practical clinical use of EHG. This work potentially contributes to the early detection of authentic preterm labor women and will allow clinicians to design individual patient strategies for maternal health surveillance systems and predict adverse pregnancy outcomes.

Regional identification of information flow termination of electrohysterographic signals: Towards understanding human uterine electrical propagation

BACKGROUND AND OBJECTIVE

The uterine electrohysterogram (EHG) contains important information about electrical signal propagation which may be useful to monitor and predict the progress of pregnancy towards parturition. Directed information processing has the potential to be of use in studying EHG recordings. However, so far, there is no directed information-based estimation scheme that has been applied to investigating the propagation of human EHG recordings. To realize this, the approach of directed information and its reliability and adaptability should be scientifically studied.

METHODS

We demonstrated an estimation scheme of directed information to identify the spatiotemporal relationship between the recording channels of EHG signal and assess the algorithm reliability initially using simulated data. Further, a regional identification of information flow termination (RIIFT) approach was developed and applied for the first time to extant multichannel EHG signals to reveal the terminal zone of propagation of the electrical activity associated with uterine contraction. RIIFT operates by estimating the pairwise directed information between neighboring EHG channels and identifying the location where there is the strongest inward flow of information. The method was then applied to publicly-available experimental data obtained from pregnant women with the use of electrodes arranged in a 4-by-4 grid.

RESULTS

Our results are consistent with the suggestions from the previous studies with the added identification of preferential sites of excitation termination - within the estimated area, the direction of surface action potential propagation towards the medial axis of uterus during contraction was discovered for 72.15% of the total cases, demonstrating that our RIIFT method is a potential tool to investigate EHG propagation for advancing our understanding human uterine excitability.

CONCLUSIONS

We developed a new approach and applied it to multichannel human EHG recordings to investigate the electrical signal propagation involved in uterine contraction. This provides an important platform for future studies to fill knowledge gaps in the spatiotemporal patterns of electrical excitation of the human uterus.

Accessing Artificial Intelligence for Fetus Health Status Using Hybrid Deep Learning Algorithm (AlexNet-SVM) on Cardiotocographic Data

Artificial intelligence is serving as an impetus in digital health, clinical support, and health informatics for an informed patient’s outcome. Previous studies only consider classification accuracies of cardiotocographic (CTG) datasets and disregard computational time, which is a relevant parameter in a clinical environment. This paper proposes a modified deep neural algorithm to classify untapped pathological and suspicious CTG recordings with the desired time complexity. In our newly developed classification algorithm, AlexNet architecture is merged with support vector machines (SVMs) at the fully connected layers to reduce time complexity. We used an open-source UCI (Machine Learning Repository) dataset of cardiotocographic (CTG) recordings. We divided 2126 CTG recordings into 3 classes (Normal, Pathological, and Suspected), including 23 attributes that were dynamically programmed and fed to our algorithm. We employed a deep transfer learning (TL) mechanism to transfer prelearned features to our model. To reduce time complexity, we implemented a strategy wherein layers in the convolutional base were partially trained to leave others in the frozen states. We used an ADAM optimizer for the optimization of hyperparameters. The presented algorithm also outperforms the leading architectures (RCNNs, ResNet, DenseNet, and GoogleNet) with respect to real-time accuracies, sensitivities, and specificities of 99.72%, 96.67%, and 99.6%, respectively, making it a viable candidate for clinical settings after real-time validation.

Recognition of cervical precancerous lesions based on probability distribution feature guidance

INTRODUCTION:

Cervical cancer is a high incidence of cancer in women and cervical precancerous screening plays an important role in reducing the mortality rate.

METHOD:

- In this study, we proposed a multichannel feature extraction method based on the probability distribution features of the acetowhite (AW) region to identify cervical precancerous lesions, with the overarching goal to improve the accuracy of cervical precancerous screening. A k-means clustering algorithm was first used to extract the cervical region images from the original colposcopy images. We then used a deep learning model called DeepLab V3+ to segment the AW region of the cervical image after the acetic acid experiment, from which the probability distribution map of the AW region after segmentation was obtained. This probability distribution map was fed into a neural network classification model for multichannel feature extraction, which resulted in the final classification performance.

RESULT:

Results of the experimental evaluation showed that the proposed method achieved an average accuracy of 87.7%, an average sensitivity of 89.3%, and an average specificity of 85.6%. Compared with the methods that did not add segmented probability features, the proposed method increased the average accuracy rate, sensitivity, and specificity by 8.3%, 8%, and 8.4%, respectively.

CONCLUSION:

Overall, the proposed method holds great promise for enhancing the screening of cervical precancerous lesions in the clinic by providing the physician with more reliable screening results that might reduce their workload.

Assessment of Features between Multichannel Electrohysterogram for Differentiation of Labors

Electrohysterogram (EHG) is a promising method for noninvasive monitoring of uterine electrical activity. The main purpose of this study was to characterize the multichannel EHG signals to distinguish between term delivery and preterm birth, as well as deliveries within and beyond 24 h. A total of 219 pregnant women were grouped in two ways: (1) term delivery (TD), threatened preterm labor (TPL) with the outcome of preterm birth (TPL_PB), and TPL with the outcome of term delivery (TPL_TD); (2) EHG recording time to delivery (TTD) ≤ 24 h and TTD > 24 h. Three bipolar EHG signals were analyzed for the 30 min recording. Six EHG features between multiple channels, including multivariate sample entropy, mutual information, correlation coefficient, coherence, direct partial Granger causality, and direct transfer entropy, were extracted to characterize the coupling and information flow between channels. Significant differences were found for these six features between TPL and TD, and between TTD ≤ 24 h and TTD > 24 h. No significant difference was found between TPL_PB and TPL_TD. The results indicated that EHG signals of TD were more regular and synchronized than TPL, and stronger coupling between multichannel EHG signals was exhibited as delivery approaches. In addition, EHG signals propagate downward for the majority of pregnant women regardless of different labors. In conclusion, the coupling and propagation features extracted from multichannel EHG signals could be used to differentiate term delivery and preterm birth and may predict delivery within and beyond 24 h.

Developing neural network model for predicting cardiac and cardiovascular health using bioelectrical signal processing

Coronary vascular disease (CHD) is one of the most fatal diseases worldwide. Cardio vascular diseases are not easily diagnosed in early disease stages. Early diagnosis is important for effective treatment, however, medical diagnoses are based on physician's personal experiences of the disease which increase time and testing cost to reach diagnosis. Physicians assess patients' condition based on electrocardiography, sonography and blood test results. In this research we develop classification model of the functional state of the cardiovascular system based on the monitoring of the evolution of the amplitudes of the first and second harmonics of the system rhythm of 0.1 Hz. We separate the signal to three streams; the first stream works with natural electro cardio signal, the other two streams are obtained as a result of frequency analysis of the amplitude- and frequency-detected electro cardio signal. We use sliding window of a demodulated electro cardio signal by means of amplitude and frequency detectors. The developed NN model showed an increase in accuracy of diagnostic efficiency by 11%. The neural network model can be trained to give accurate early detection of disease class.

IDENTIFICATION OF UTERINE CONTRACTIONS BY AN ENSEMBLE OF GAUSSIAN PROCESSES

Identifying uterine contractions with the aid of machine learning methods is necessary vis-á-vis their use in combination with fetal heart rates and other clinical data for the assessment of a fetus wellbeing. In this paper, we study contraction identification by processing noisy signals due to uterine activities. We propose a complete four-step method where we address the imbalanced classification problem with an ensemble Gaussian process classifier, where the Gaussian process latent variable model is used as a decision-maker. The results of both simulation and real data show promising performance compared to existing methods.

Optimization of Imminent Labor Prediction Systems in Women with Threatened Preterm Labor Based on Electrohysterography

Preterm birth is the leading cause of death in newborns and the survivors are prone to health complications. Threatened preterm labor (TPL) is the most common cause of hospitalization in the second half of pregnancy. The current methods used in clinical practice to diagnose preterm labor, the Bishop score or cervical length, have high negative predictive values but not positive ones. In this work we analyzed the performance of computationally efficient classification algorithms, based on electrohysterographic recordings (EHG), such as random forest (RF), extreme learning machine (ELM) and K-nearest neighbors (KNN) for imminent labor (<7 days) prediction in women with TPL, using the 50th or 10th–90th percentiles of temporal, spectral and nonlinear EHG parameters with and without obstetric data inputs. Two criteria were assessed for the classifier design: F1-score and sensitivity. RF_{F1_2} and ELM_{F1_2} provided the highest F1-score values in the validation dataset, (88.17 ± 8.34% and 90.2 ± 4.43%) with the 50th percentile of EHG and obstetric inputs. ELM_{F1_2} outperformed RF_{F1_2} in sensitivity, being similar to those of ELM_Sens (sensitivity optimization). The 10th–90th percentiles did not provide a significant improvement over the 50th percentile. KNN performance was highly sensitive to the input dataset, with a high generalization capability.

Performance comparison of coupling-evaluation methods in discriminating between pregnancy and labor EHG signals

BACKGROUND

Recent years have seen an increased interest in electrohysterogram (EHG) signals as a means to evaluate the synchronization of uterine contractions. Several studies have pointed out that the quality of signal processing - and hence the interpretation of measurement results - is affected significantly by the choice of measurement technique and the presence of non-stationary frequency content in EHG signals. To our knowledge, the effect of time variance on the quality of EHG signal processing has never been fully investigated. How best to process EHG signals with the goal of distinguishing labor-induced contractions from their harmless, pre-labor cousins, remains an open question.

METHOD

Our methodology is based on three pillars. The first consists of a new method for EHG preprocessing in which we apply a second-order Butterworth filter to retain only the EHG fast-wave, low-frequency band (FWL), then use a bivariate piecewise stationary pre-segmentation (bPSP) algorithm to segment the EHG signal into stationary parts. The second pillar addresses the estimation of connectivity and directionality using three methods: nonlinear correlation coefficient (h²), general synchronization (H), and Granger causality (GC). The third pillar is related to signal classification and discrimination between pregnancy and labor using receiver operating curves (ROC) and connectivity and direction maps. For this purpose, we analyze the impact of four factors on data processing efficiency: i) method of connectivity detection, ii) effect of piecewise stationary segmentation preprocessing, iii) retained frequency content and iv) electrode configuration used for EHG recording (bipolar vs. unipolar).

RESULTS

Our results show that piecewise signal segmentation and filtering considerably improves classification performance and statistical significance for some connectivity methods, in particular the h². To this end we propose a new approach (detailed below) for h² called Filtered-Windowed (FW) h² that better highlights the differences between pregnancy and labor in the connectivity matrix and directionality maps.

CONCLUSIONS

This is the first comparative study of the effects of multiple processing factors on connectivity measurement efficiency. Our results indicate that appropriate preprocessing can improve the differentiation of pregnancy and labor-induced contraction signals and may lead to innovative applications in the prevention of preterm labor.

Automatic recognition of uterine contractions with electrohysterogram signals based on the zero-crossing rate

Uterine contraction (UC) is an essential clinical indicator in the progress of labour and delivery. Electrohysterogram (EHG) signals recorded on the abdomen of pregnant women reflect the uterine electrical activity. This study proposes a novel algorithm for automatic recognition of UCs with EHG signals to improve the accuracy of detecting UCs. EHG signals by electrodes, the tension of the abdominal wall by tocodynamometry (TOCO) and maternal perception were recorded simultaneously in 54 pregnant women. The zero-crossing rate (ZCR) of the EHG signal and its power were calculated to modulate the raw EHG signal and highlight the EHG bursts. Then the envelope was extracted from the modulated EHG for UC recognition. Besides, UC was also detected by the conventional TOCO signal. Taking maternal perception as a reference, the UCs recognized by EHG and TOCO were evaluated with the sensitivity, positive predictive value (PPV), and UC parameters. The results show that the sensitivity and PPV are 87.8% and 93.18% for EHG, and 84.04% and 90.89% for TOCO. EHG detected a larger number of UCs than TOCO, which is closer to maternal perception. The duration and frequency of UC obtained from EHG and TOCO were not significantly different (p > 0.05). In conclusion, the proposed UC recognition algorithm has high accuracy and simple calculation which could be used for real-time analysis of EHG signals and long-term monitoring of UCs.

Detection of Epileptic Seizure Using Pretrained Deep Convolutional Neural Network and Transfer Learning

INTRODUCTION

The diagnosis of epilepsy takes a certain process, depending entirely on the attending physician. However, the human factor may cause erroneous diagnosis in the analysis of the EEG signal. In the past 2 decades, many advanced signal processing and machine learning methods have been developed for the detection of epileptic seizures. However, many of these methods require large data sets and complex operations.

METHODS

In this study, an end-to-end machine learning model is presented for detection of epileptic seizure using the pretrained deep two-dimensional convolutional neural network (CNN) and the concept of transfer learning. The EEG signal is converted directly into visual data with a spectrogram and used directly as input data.

RESULTS

The authors analyzed the results of the training of the proposed pretrained AlexNet CNN model. Both binary and ternary classifications were performed without any extra procedure such as feature extraction. By performing data set creation from short-term spectrogram graphic images, the authors were able to achieve 100% accuracy for binary classification for epileptic seizure detection and 100% for ternary classification.

DISCUSSION/CONCLUSION

The proposed automatic identification and classification model can help in the early diagnosis of epilepsy, thus providing the opportunity for effective early treatment.

Diagnosis of cervical precancerous lesions based on multimodal feature changes

To realize the automatic diagnosis of cervical intraepithelial neoplasia (CIN) cases by preacetic acid test and postacetic acid test colposcopy images, this paper proposes a method of cervical precancerous lesion diagnosis based on multimodal feature changes. First, the preacetic acid test and postacetic acid test colposcopy images were registered based on cross-correlation and projection transformation, and then the cervical region was extracted by the k-means clustering algorithm. Finally, a deep learning network was used to extract features and classify the preacetic acid test and postacetic acid test cervical images after registration. Finally, the proposed method achieves a classification accuracy of 86.3%, a sensitivity of 84.1%, and a specificity of 89.8% in 60 test cases. Experimental results show that this method can make better use of the multimodal features of colposcopy images and has lower requirements for medical staff in the process of data acquisition. It has certain clinical significance in cervical cancer precancerous lesion screening systems.

Analysis of Electrohysterographic Signal Propagation Direction during Uterine Contraction: the Application of Directed Information

The potential of using the information of uterine contractions (UCs) derived from electrohysterogram (EHG) has been recognized in early detection of preterm delivery. A better understanding of the conduction property of EHG is clinically useful for developing advanced methods to achieve a reliable prediction of preterm delivery. In this paper, a method to analyze the destination of EHG propagation has been proposed via the estimation of directed information (DI) between each pair of neighboring channels with a novel propagation terminal zone (PTZ) identification algorithm. The proposed method was applied to experimental data from the Icelandic 16-electrode EHG database. The results demonstrated that for more than 81.8% participants, the PTZ was identified along the medial axis of uterus, among which more than half have their PTZ determined in the center between the uterine fundus and public symphysis, which indicated a great probability of propagation of EHG signals towards the center of uterus plane.Clinical relevance- This study makes a fundamental contribution for predicting preterm delivery, which can provide improvement in obstetric care towards pregnancy monitoring.

COVID19XrayNet: A Two-Step Transfer Learning Model for the COVID-19 Detecting Problem Based on a Limited Number of Chest X-Ray Images

The novel coronavirus severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused a major pandemic outbreak recently. Various diagnostic technologies have been under active development. The novel coronavirus disease (COVID-19) may induce pulmonary failures, and chest X-ray imaging becomes one of the major confirmed diagnostic technologies. The very limited number of publicly available samples has rendered the training of the deep neural networks unstable and inaccurate. This study proposed a two-step transfer learning pipeline and a deep residual network framework COVID19XrayNet for the COVID-19 detection problem based on chest X-ray images. COVID19XrayNet firstly tunes the transferred model on a large dataset of chest X-ray images, which is further tuned using a small dataset of annotated chest X-ray images. The final model achieved 0.9108 accuracy. The experimental data also suggested that the model may be improved with more training samples being released.

A wavelet-based algorithm for automated analysis of external tocography: How does it compare to human interpretation?

BACKGROUND

Studies which use external tocography to explore the relationship between increased intrapartum uterine activity and foetal outcomes are feasible because the technology is safe and ubiquitous. However, periods of poor signal quality are common. We developed an algorithm which aims to calculate tocograph summary variables based on well-recorded contractions only, ignoring artefact and excluding sections deemed uninterpretable. The aim of this study was to test that algorithm's reliability.

METHODS

Whole recordings from labours at ≥35 weeks of gestation were randomly selected without regard to quality. Contractions and rest intervals were measured by two humans independently, and by the algorithm using two sets of models; one based on a series of pre-defined thresholds, and another trained to imitate one of the human interpreters. The absolute agreement intraclass correlation coefficient (ICC) was calculated using a two-way random effects model.

RESULTS

The training dataset included data from 106 tocographs. Of the tested algorithms, AdaBoost showed the highest initial cross-validated accuracy and proceeded to optimization. Forty tocographs were included in the validation set. The ICCs for the per tocograph mean contraction rates were; human B to human A: 0.940 (0.890-0.968), human A to initial models: 0.944 (0.898-0.970), human A to trained models 0.962 (0.927-0.980), human B to initial models: 0.930 (0.872-0.962), human B to trained models: 0.948 (0.903-0.972).

CONCLUSIONS

The algorithm described approximates interpretation of external tocography performed by trained humans. The performance of the AdaBoost trained models was marginally superior compared to the initial models.

Acquisition and Analysis of Electrohysterogram Signal

Electrohysterogram (EHG) signal is the signal related to action potentials propagating through smooth muscle cells of the uterus (myometrium) to the abdomen of pregnant women which is also known as uterine contraction signal. Cardiotocography (CTG) is the most common method used for monitoring fetal heart rate (FHR) and uterine contractions during pregnancy and labor. This method detects mechanical activity of fetal heart and uterus, however, it provides low accuracy and sensibility and hence more accurate methods are required. The abdominal electrode method of FECG monitoring and Electrohysterography (EHG) are alternative noninvasive method to monitor the FHR and uterine contractions during pregnancy which provides better results compared to CTG. Each information such as the frequency of uterine contractions, length of the contraction and contraction power of uterus, indicates the condition of the uterus which will help the obstetricians to identify the progress of labor. All these above mentioned parameters can be identified from the EHG signal acquired non-invasively by placing the electrodes on the abdomen of the pregnant women. In this work the acquisition of EHG signal as well as analysis of EHG signal in both antepartum condition and labor condition have been carried out and parameters such as number of contractions, contraction duration, amplitude, power of contraction are computed and the quantitative analysis of EHG signals in both above mentioned conditions are performed and it is compared with the simultaneously recorded uterine contraction signal parameters from Cardiotocography (CTG).

Evaluation of electrohysterogram measured from different gestational weeks for recognizing preterm delivery: a preliminary study using random Forest

Developing a computational method for recognizing preterm delivery is important for timely diagnosis and treatment of preterm delivery. The main aim of this study was to evaluate electrohysterogram (EHG) signals recorded at different gestational weeks for recognizing the preterm delivery using random forest (RF). EHG signals from 300 pregnant women were divided into two groups depending on when the signals were recorded: i) preterm and term delivery with EHG recorded before the 26^th week of gestation (denoted by PE and TE group), and ii) preterm and term delivery with EHG recorded during or after the 26^th week of gestation (denoted by PL and TL group). 31 linear features and nonlinear features were derived from each EHG signal, and then compared comprehensively within PE and TE group, and PL and TL group. After employing the adaptive synthetic sampling approach and six-fold cross-validation, the accuracy (ACC), sensitivity, specificity and area under the curve (AUC) were applied to evaluate RF classification. For PL and TL group, RF achieved the ACC of 0.93, sensitivity of 0.89, specificity of 0.97, and AUC of 0.80. Similarly, their corresponding values were 0.92, 0.88, 0.96 and 0.88 for PE and TE group, indicating that RF could be used to recognize preterm delivery effectively with EHG signals recorded before the 26^th week of gestation.

Preliminary Study on the Efficient Electrohysterogram Segments for Recognizing Uterine Contractions with Convolutional Neural Networks

Background

Uterine contraction (UC) is the tightening and shortening of the uterine muscles which can indicate the progress of pregnancy towards delivery. Electrohysterogram (EHG), which reflects uterine electrical activities, has recently been studied for UC monitoring. In this paper, we aimed to evaluate different EHG segments for recognizing UCs using the convolutional neural network (CNN).

Materials and Methods

In the open-access Icelandic 16-electrode EHG database (122 recordings from 45 pregnant women), 7136 UC and 7136 non-UC EHG segments with the duration of 60 s were manually extracted from 107 recordings of 40 pregnant women to develop a CNN model. A fivefold cross-validation was applied to evaluate the CNN based on sensitivity (SE), specificity (SP), and accuracy (ACC). Then, 1056 UC and 1056 non-UC EHG segments were extracted from the other 15 recordings of 5 pregnant women. Furthermore, the developed CNN model was applied to identify UCs using different EHG segments with the durations of 10 s, 20 s, and 30 s.

Results

The CNN achieved the average SE, SP, and ACC of 0.82, 0.93, and 0.88 for a 60 s EHG segment. The EHG segments of 10 s, 20 s, and 30 s around the TOCO peak achieved higher SE and ACC than the other segments with the same duration. The values of SE from 20 s EHG segments around the TOCO peak were higher than those from 10 s to 30 s EHG segments on the same side of the TOCO peak.

Conclusion

The proposed method could be used to determine the efficient EHG segments for recognizing UC with the CNN.