Classification performance of the frequency-related parameters derived from uterine EMG signals

A Novel, Cardiac-Derived Algorithm for Uterine Activity Monitoring in a Wearable Remote Device

Background: Uterine activity (UA) monitoring is an essential element of pregnancy management. The gold-standard intrauterine pressure catheter (IUPC) is invasive and requires ruptured membranes, while the standard-of-care, external tocodynamometry (TOCO)’s accuracy is hampered by obesity, maternal movements, and belt positioning. There is an urgent need to develop telehealth tools enabling patients to remotely access care. Here, we describe and demonstrate a novel algorithm enabling remote, non-invasive detection and monitoring of UA by analyzing the modulation of the maternal electrocardiographic and phonocardiographic signals. The algorithm was designed and implemented as part of a wireless, FDA-cleared device designed for remote pregnancy monitoring. Two separate prospective, comparative, open-label, multi-center studies were conducted to test this algorithm.

Methods: In the intrapartum study, 41 laboring women were simultaneously monitored with IUPC and the remote pregnancy monitoring device. Ten patients were also monitored with TOCO. In the antepartum study, 147 pregnant women were simultaneously monitored with TOCO and the remote pregnancy monitoring device.

Results: In the intrapartum study, the remote pregnancy monitoring device and TOCO had sensitivities of 89.8 and 38.5%, respectively, and false discovery rates (FDRs) of 8.6 and 1.9%, respectively. In the antepartum study, a direct comparison of the remote pregnancy monitoring device to TOCO yielded a sensitivity of 94% and FDR of 31.1%. This high FDR is likely related to the low sensitivity of TOCO.

Conclusion: UA monitoring via the new algorithm embedded in the remote pregnancy monitoring device is accurate and reliable and more precise than TOCO standard of care. Together with the previously reported remote fetal heart rate monitoring capabilities, this novel method for UA detection expands the remote pregnancy monitoring device’s capabilities to include surveillance, such as non-stress tests, greatly benefiting women and providers seeking telehealth solutions for pregnancy care.

Electrodes in external electrohysterography: a systematic literature review

BACKGROUND

In low-income countries, pregnant women do not have easy access to health care, especially in rural and peri-urban areas. In this context, they can be surprised by the uterine contractions that precede childbirth and sometimes find themselves giving birth at home or on the way to the nearest health facility (located miles away from their home). In view of the development of an external uterine electrohysterogram acquisition system for labour prediction, a review of the literature on electrodes and their characteristics is necessary.

METHODS

A comprehensive literature review was conducted to collate information on the use of electrodes in external EHG recording and their characteristics.

RESULTS

Wet electrodes based on Ag/AgCl redox chemistry are the most common type of electrodes for EHG, employed in different configurations on the pregnant woman's abdomen. All positioning configurations are around the vertical median axis if they are not placed directly on it. Positioning below the navel seems to be the most efficient. The number of source, reference, and ground electrodes used varies from one author to another, as does the distance between the electrodes.

CONCLUSION

Two well-positioned source electrodes on the vertical median axis, with ground electrode on the right side of the hip and reference one on the left side, are able to generate a good external EHG recording signal. The minimum allowed inter-electrode distance is approximately 17.5 to 25mm.

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.

Comparison of electrohysterogram characteristics during uterine contraction and non-contraction during labor

Uterine contraction is one of the most important indication in the labor progression. Electrohysterogram (EHG) is a promising method for monitoring uterine contraction and discriminating efficient and inefficient contractions. This study aims to analyze the difference of EHG signals between two groups. EHG signals are recorded with abdominal electrodes from 20 pregnant women, including 10 in term labor group and 10 in non-labor group. Typical linear and nonlinear characteristics of EHG signals, including root mean square (RMS), peak frequency (PF), median frequency (MDF), mean frequency (MNF), parameters from wavelet decomposition (W4, W5) and time reversibility (Tr) are extracted. These characteristics are compared between contraction and non-contraction in term labor group and non-labor group. The result shows that RMS, W4 and W5 of contraction are significantly larger than non-contraction both within term labor group and between two groups (all p<;0.001). However, MDF and MNF are significantly smaller (all p<;0.05). Furthermore, all characteristics of non-contraction show no significant difference between two groups, except MNF. The variability of RMS, W4, W5 and Tr of contraction are significantly larger than non-contraction both within term labor group and between two groups (all p<;0.05, with p<;0.001 for W5 and Tr). However, the variability of MDF, PF and MNF are significantly smaller (all p<;0.05). Moreover, the variability of all characteristics of non-contraction shows no significant difference between two groups, except MNF. We have shown that characteristics of EHG signals and their variability during contraction are quite different from non-contraction. Therefore, it is feasible to separate uterine contractions and monitor uterine activity with EHG signals.

The Icelandic 16-electrode electrohysterogram database

External recordings of the electrohysterogram (EHG) can provide new knowledge on uterine electrical activity associated with contractions. Better understanding of the mechanisms underlying labor can contribute to preventing preterm birth which is the main cause of mortality and morbidity in newborns. Promising results using the EHG for labor prediction and other uses in obstetric care are the drivers of this work. This paper presents a database of 122 4-by-4 electrode EHG recordings performed on 45 pregnant women using a standardized recording protocol and a placement guide system. The recordings were performed in Iceland between 2008 and 2010. Of the 45 participants, 32 were measured repeatedly during the same pregnancy and participated in two to seven recordings. Recordings were performed in the third trimester (112 recordings) and during labor (10 recordings). The database includes simultaneously recorded tocographs, annotations of events and obstetric information on participants. The publication of this database enables independent and novel analysis of multi-electrode EHG by the researchers in the field and hopefully development towards new life-saving technology.

Automatic epileptic seizure detection using scalp EEG and advanced artificial intelligence techniques

The epilepsies are a heterogeneous group of neurological disorders and syndromes characterised by recurrent, involuntary, paroxysmal seizure activity, which is often associated with a clinicoelectrical correlate on the electroencephalogram. The diagnosis of epilepsy is usually made by a neurologist but can be difficult to be made in the early stages. Supporting paraclinical evidence obtained from magnetic resonance imaging and electroencephalography may enable clinicians to make a diagnosis of epilepsy and investigate treatment earlier. However, electroencephalogram capture and interpretation are time consuming and can be expensive due to the need for trained specialists to perform the interpretation. Automated detection of correlates of seizure activity may be a solution. In this paper, we present a supervised machine learning approach that classifies seizure and nonseizure records using an open dataset containing 342 records. Our results show an improvement on existing studies by as much as 10% in most cases with a sensitivity of 93%, specificity of 94%, and area under the curve of 98% with a 6% global error using a k-class nearest neighbour classifier. We propose that such an approach could have clinical applications in the investigation of patients with suspected seizure disorders.

Prediction of preterm deliveries from EHG signals using machine learning

There has been some improvement in the treatment of preterm infants, which has helped to increase their chance of survival. However, the rate of premature births is still globally increasing. As a result, this group of infants are most at risk of developing severe medical conditions that can affect the respiratory, gastrointestinal, immune, central nervous, auditory and visual systems. In extreme cases, this can also lead to long-term conditions, such as cerebral palsy, mental retardation, learning difficulties, including poor health and growth. In the US alone, the societal and economic cost of preterm births, in 2005, was estimated to be $26.2 billion, per annum. In the UK, this value was close to £2.95 billion, in 2009. Many believe that a better understanding of why preterm births occur, and a strategic focus on prevention, will help to improve the health of children and reduce healthcare costs. At present, most methods of preterm birth prediction are subjective. However, a strong body of evidence suggests the analysis of uterine electrical signals (Electrohysterography), could provide a viable way of diagnosing true labour and predict preterm deliveries. Most Electrohysterography studies focus on true labour detection during the final seven days, before labour. The challenge is to utilise Electrohysterography techniques to predict preterm delivery earlier in the pregnancy. This paper explores this idea further and presents a supervised machine learning approach that classifies term and preterm records, using an open source dataset containing 300 records (38 preterm and 262 term). The synthetic minority oversampling technique is used to oversample the minority preterm class, and cross validation techniques, are used to evaluate the dataset against other similar studies. Our approach shows an improvement on existing studies with 96% sensitivity, 90% specificity, and a 95% area under the curve value with 8% global error using the polynomial classifier.