Linear Decomposition Research Articles

Using geometric mode decomposition for the background noise suppression on microseismic data

AbstractMicroseismic datasets typically have relatively low signal‐to‐noise ratio waveforms. To that end, several noise suppression techniques are often applied to improve the signal‐to‐noise ratio of the recorded waveforms. We apply a linear geometric mode decomposition approach to microseismic datasets for background noise suppression. The geometric mode decomposition method optimizes linear patterns within amplitude–frequency modulated modes and can efficiently distinguish microseismic events (signal) from the background noise. This method can also split linear and non‐linear dispersive seismic events into linear modes. The segmented events in different modes can then be added carefully to reconstruct the denoised signal. The application of geometric mode decomposition is well suited for microseismic acquisitions with smaller receiver spacing, where the signal may exhibit either (nearly) linear or non‐linear recording patterns, depending on the source location relative to the receiver array. Using synthetic and real microseismic data examples from limited‐aperture downhole recordings only, we show that geometric mode decomposition is robust in suppressing the background noise from the recorded waveforms. We also compare the filtering results from geometric mode decomposition with those obtained from FX‐Decon and one‐dimensional variational mode decomposition methods. For the examples used, geometric mode decomposition outperforms both FX‐Decon and one‐dimensional variational mode decomposition in background noise suppression.

Frequency dependence of coordinated pupil and eye movements for binocular disparity tracking.

Coordinated alignment of the eyes during gaze fixation and eye movements are an important component of normal visual function. We have previously described the coordinated behavior of convergence eye movements and pupillary responses using a 0.1 Hz binocular disparity-driven sine profile and a step profile. The goal of this publication is to further characterize ocular vergence-pupil size coordination over a wider range of frequencies of ocular disparity stimulation in normal subjects. Binocular disparity stimulation is generated by presentation of independent targets to each eye on a virtual reality display, while eye movements and pupil size are measured by an embedded video-oculography system. This design allows us to study two complimentary analyses of this motion relationship. First, a macroscale analysis describes the vergence angle of the eyes in response to binocular disparity target movement and pupil area as a function of the observed vergence response. Second, a microscale analysis performs a piecewise linear decomposition of the vergence angle and pupil relationship to permit more nuanced findings. These analyses identified three main features of controlled coupling of pupil and convergence eye movements. First, a near response relationship operates with increasing prevalence during convergence (relative to the "baseline" angle); the coupling is higher with increased convergence in this range. Second, the prevalence of "near response"-type coupling decreases monotonically in the diverging direction; the decrease persists after the targets move (converge back) from maximum divergence toward the baseline positions, with a minimum prevalence of near response segments near the baseline target position. Third, an opposite polarity pupil response is infrequent, but tends to be more prevalent when the vergence angles are at maximum convergence or divergence for a sinusoidal binocular disparity task. We suggest that the latter response is an exploratory "range-validation" when binocular disparity is relatively constant. In a broader sense, these findings describe operating characteristics of the near response in normal subjects and form a basis for quantitative assessments of function in conditions such as convergence insufficiency and mild traumatic brain injury.

A Variational Ansatz for Taylorized Imaginary Time Evolution.

Being able to predict molecular properties and interactions is of utmost interest for academia as well as industry. But the vast complexity of strongly correlated molecular systems limits the performance of classical algorithms. In contrast, quantum computation has the potential to be a game changer in the field of molecular simulations. Despite the hope in quantum computation, the capabilities of current quantum computers are still insufficient for handling molecular systems of interest. In this paper, we propose a variational ansatz for today's noisy quantum computers to calculate the ground state with the help of imaginary time evolution. Although the imaginary time evolution operator is not unitary, it can be implemented on a quantum computer by a linear decomposition and subsequent Taylor series expansion. This has the advantage that only a set of shallow circuits needs to be computed on a quantum computer. The parallel nature of this algorithm can be exploited to speed-up simulations even further, if a privileged access to quantum computers is granted.

Optimising for Interpretability: Convolutional Dynamic Alignment Networks.

We introduce a new family of neural network models called Convolutional Dynamic Alignment Networks (CoDA Nets), which are performant classifiers with a high degree of inherent interpretability. Their core building blocks are Dynamic Alignment Units (DAUs), which are optimised to transform their inputs with dynamically computed weight vectors that align with task-relevant patterns. As a result, CoDA Nets model the classification prediction through a series of input-dependent linear transformations, allowing for linear decomposition of the output into individual input contributions. Given the alignment of the DAUs, the resulting contribution maps align with discriminative input patterns. These model-inherent decompositions are of high visual quality and outperform existing attribution methods under quantitative metrics. Further, CoDA Nets constitute performant classifiers, achieving on par results to ResNet and VGG models on e.g. CIFAR-10 and TinyImagenet. Lastly, CoDA Nets can be combined with conventional neural network models to yield powerful classifiers that more easily scale to complex datasets such as Imagenet whilst exhibiting an increased interpretable depth, i.e., the output can be explained well in terms of contributions from intermediate layers within the network.

Declining trend in cigarette smoking among U.S. adults over 2008–2018: A decomposition analysis

The United States (U.S.) witnessed considerable reduction in cigarette smoking prevalence in the recent past. While the correlates of smoking prevalence and related disparities among U.S. adults are well documented, there is limited information on how this success was shared among different population sub-groups. Based on data from the National Health Interview Surveys, 2008 and 2018, representative of non-institutionalized U.S. adults (18 years and above), we applied the threefold Kitawaga-Oaxaca-Blinder linear decomposition analysis. We decomposed the trends in cigarette smoking prevalence, smoking initiation, and successful cessation into changes in population characteristics holding smoking propensities constant (compositional change), changes in smoking propensities by population characteristics holding population composition constant (structural change), and the unmeasured macro-level changes affecting smoking behavior in different population sub-groups at differential rates (residual change) to quantify the shares of population sub-groups by sex, age, race/ethnicity, education, marital status, employment status, health insurance coverage, family income, and region of residence in the overall change in smoking rates. The analysis shows that decreases in smoking propensities regardless of the changes in population composition accounted for 66.4% of the reduction in smoking prevalence and 88.7% of the reduction in smoking initiation. The major reductions in smoking propensity were among Medicaid recipients and young adults (ages 18–24 years). The 25-44-year-olds experienced moderate increase in successful smoking cessation, while the overall successful smoking cessation rate remained steady. Taken together, consistent reduction in smoking among U.S. adults by all major population characteristics, accompanied by disproportionately larger reduction in smoking propensities among the population sub-groups with initially higher smoking propensity compared to the national average, characterized the decline in overall cigarette smoking. Strengthening proven tobacco control measures with targeted interventions to reduce smoking propensities among underserved populations is key to continued success in reducing smoking overall and remedying inequities in smoking and population health.

Two-dimensional complex wavelet transform for linear noise attenuation and image decomposition

Abstract For developing a high-fidelity, high-resolution seismic denoising method, we use the two-dimensional complex wavelet transform (2D CWT) to analyze noise and signals. By investigating a surface wave's features and evaluating factors affecting the fidelity of the method, the best practice for the wavelet transform-based denoising has been established. First, static and normal moveout correction are applied on shot gathers. Then, 2D CWT is used to attenuate linear noises. The results demonstrate that the proposed method and practice significantly attenuate noises and preserve the signal's amplitudes and frequency band. In addition to denoising, we also apply the 2D CWT to decompose a seismic image into multiscale images with different resolutions. Multiscale decomposed images derive more detailed information for subsurface structures and fault networks. The decomposed images depict sharper structures and reveal detailed features of faults more significantly than the original images.

L-SHADE with parameter decomposition for photovoltaic modules parameter identification under different temperature and irradiance

The performance of photovoltaic (PV) system relies on the accurate determination of unknown parameters in PV models, such as photo-generated current, diode current, and resistance. These unknown parameters may vary with different temperature and irradiance conditions, which greatly increases the difficulty of identifying these unknown parameters. Although some parameter identification methods have been proposed to solve such a problem, the accuracy and reliability of solutions obtained by these methods suffers from great challenges when temperature and irradiance are changing. In this paper, a simple and effective approach called success-history adaptation differential evolution with linear population size reduction (L-SHADE) and decomposition referred as (L-SHADED), is presented to accurately and reliably identify the unknown parameters of PV models under different temperature and irradiance. In L-SHADED, firstly, an unknown parameters decomposition technique is used to reduce the complexity of the problem. Then an advanced evolutionary algorithm L-SHADE is employed to identify the optimal unknown parameter values. The performance of proposed L-SHADED has been investigated by testing two single-diode-based PV modules (multi-crystalline KC200GT and mono-crystalline SM55) under different temperature and irradiance. The experimental comparison results demonstrate that proposed L-SHADED is almost 10 times smaller in RMSE than other methods. Furthermore, the coefficient of determination of L-SHADED reaches 1.0 in almost all conditions, which indicates that the parameters identified by L-SHADED are quite accurate.

Explaining adult obesity, severe obesity, and BMI: Five decades of change

Obesity rates have increased across all segments of society since the late 1970s, but the reason behind population-level increases in body weight remains unclear. We used the 1971–2020 NHANES data to examine whether the observed trend in obesity prevalence is attributable to changing public health behaviors (i.e., intracohort change) or changing publics (i.e., cohort replacement). We partitioned total change in mean BMI, and rates of obesity and severe obesity, into its IC and CR components using linear and algebraic decomposition methods. We found that the IC mechanism (i.e., broad sectors of individuals changing) plays a dominant role in the overall increase in mean BMI, and obesity and severe obesity prevalence. Birth cohort membership (i.e., the CR mechanism) is also influencing mean BMI, and rates of obesity and severe obesity, but in differing ways. Specifically, the large positive IC and the small positive CR effects are amplifying one another, thus creating a steep increase in the observed rates of severe obesity. Conversely, the large positive IC effect is offset by a small negative CR effect, which created a more gradual rise in mean BMI and rates of obesity. Furthermore, we computed total change for models that entered separately sociodemographic, lifestyle, nutritional, and physical activity measures to estimate differences in mean BMI, and rates of obesity and severe obesity, among cohorts and time periods. Adjustment for all the compositional differences among the cohorts during the study period indicate that a combination of a more pronounced IC and a less pronounced CR drove the observed increase in mean BMI, and rates of obesity and severe obesity. Thus, “universal prevention” (i.e., entire community) strategies for healthy weight promotion may need to be combined with “selective prevention” (i.e., at-risk groups) and/or “targeted prevention” (i.e., at-risk individuals) approaches in order to reverse the obesity epidemic.

One-step fluorometric determination of multiple-component dissolved organic matter in aquatic environments

Dissolved organic matter (DOM) is ubiquitous in aqueous environments and is composed of different components that play different but important roles in the migration and the fate of pollutants, emergence of the disinfect byproduct, thus requiring quantitative characterization. However, until now, simultaneous quantification of the main contents in DOM, i.e., saccharides, proteins, and humic substances, has been difficult, impeding us from understanding and predicting the environmental behaviors of typical pollutants. In this work, a fluorescence approach based on the excitation emission matrix (EEM), combined with a new algorithm, denoted matrix reconstruction coupled with prior linear decomposition (MR-PLD), was developed to quantify multiple DOM simultaneously. First, a set of simulated water samples consisting of glucose, tryptones, and humic acid (HA) were analyzed using MR-PLD to validate the feasibility of the method. The DOM components could be reliably determined with a higher accuracy than parallel factor analysis (PARAFAC) and Parallel Factor Framework-Linear Regression (PFFLR), also with a more convenient procedure than conventional PLD. Second, both actual simulated and experimental methods were performed to test the anti-interference performance of MR-PLD, indicating that the quantification of DOM would not be significantly impacted by other fluorophores. Finally, several actual water samples from natural waters and wastewater treatment plants were also analyzed to confirm the robustness of this method in actual aqueous environments. This study provides a new approach to characterize DOM with EEM, contributing to its convenient concentration monitoring and the further exploration of the environmental impacts.

On the prediction of premature births in Hispanic labour patients using uterine contractions, heart beat signals and prediction machines.

Preterm birth is a global epidemic affecting millions of mothers across different ethnicities. The cause of the condition remains unknown but has recognised health-based implications, in addition to financial and economic ones. Machine Learning methods have enabled researchers to combine datasetsusinguterine contraction signals with various forms of prediction machines to improve awareness of the likelihood of premature births. This work investigates the feasibility of enhancing these prediction methods using physiological signals including uterine contractions, and foetal and maternal heart rate signals, for a population of south American women in active labour. As part of this work, the use of the Linear Series Decomposition Learner (LSDL) was seen to lead to an improvement in the prediction accuracies of all models, which included supervised and unsupervised learning models. The results from the supervised learning models showed high prediction metrics upon the physiological signals being pre-processed by the LSDL for all variations of the physiological signals. The unsupervised learning models showed good metrics for the partitioning of Preterm/Term labour patients from their uterine contraction signals but produced a comparatively lower set of results for the various kinds of heart rate signals investigated.

Pengurangan Bullwhip Effect pada Rantai Pasok Air Minum Dalam Kemasan Menggunakan Metode Collaborative Planning, Forecasting, and Replenishment (CPFR): Studi Kasus

CV Tirta Alam Dodu is a privately owned company engaged in the beverage industry in the form of AMDK. The problem faced by the company is that there is a difference between demand and sales, or better known as the bullwhip effect problem. The purpose of this study is to reduce the bullwhip effect on the supply chain and optimize the amount of product inventory according to the amount of demand. The method used is the Collaborative Planning, Forecasting, and Replenishment (CPFR) method. Based on the results of the research conducted, it can be concluded that the analysis of demand forecasting calculations on AMDK products at CV Tirta Alam Dodu was carried out using historical data on product demand through five analytical methods, including Linear Trend, Moving Average, Exponential Smoothing, Decomposition, and Winter's Model. The best forecasting resulted in Winter's Model which showed the smallest error value compared to the other four models. The way to minimize so that it does not become a bullwhip effect is to use the CPFR method, where the bullwhip effect value is obtained on glass mineral water products and gallons of mineral water of 0,985 and 0,898, respectively. This result shows that there is no amplification of demand and has reached the company's target, which is to fall below number one. The way to optimize inventory so that there is no excess stock or shortage of stock in the product is to make the amount of inventory based on the amount of demand from the forecasting results plus the safety stock.

AN EFFICIENT HYBRID METHOD TO PREDICT WIND SPEED BASED ON LINEAR REGRESSION AND VMD

To effectively improve the power dispatching, the prediction accuracy of wind power has been the concern of many scholars for many years. The wind power prediction problem is actually equivalent to the wind speed prediction problem. Based on linear regression (LR) and variational mode decomposition (VMD), in this paper, we proposed an efficient hybrid method to predict wind speed. In the proposed method, the VMD is used to decompose the signal of wind speed into several sub-signal. Compared with the original wind-speed series, each sub-signal is a more stable subsequence signal. Then, we used the LR method to predict each subsequence signal. Eventually, we obtain the final prediction results of the original wind speed series merged the forecasting values of all subsequences signal. We selected two data to test our proposed method in our experiment. Compared with several comparison methods, we found that our proposed methods has better prediction performance than other methods from the experimental results.

Hybrid DLDT algorithm based tumor classification of mammogram images

Despite the fact that current classical image classification techniques have been widely used to address real-world problems, these applications are frequently hampered by problems including disappointing results, poor classification accuracy, and constrained adaptability. This method separates the mammography image tumour feature extraction and classification processes. The effective way to improve the accuracy of mammography image tumour classification is to integrate the feature extraction and classification processes using the potent deep learning model. The deep structural advantages of multilayer nonlinear mapping and the tumour representation of well-multidimensional data linear decomposition, however, are fully utilised in this paper, which also introduces the idea of tumour representation based on shape-based feature extraction into the architecture of the deep learning network.

Nonlinear set‐membership state estimation based on the Koopman operator

SummaryIn this study, the Koopman operator is applied to solve the challenging nonlinear set‐membership (SM) state estimation problem. The basic idea is to lift the nonlinear system into a linear one with a higher dimension, and then linear SM estimation methods can be adopted. The Koopman operator is an infinite‐dimensional linear decomposition of nonlinear dynamics. This linearized system can fully describe the state evolution of the nonlinear dynamics with input, output, and noises. To apply SM state estimation algorithms, a finite‐dimensional approximation is obtained by using the data‐driven method. Meanwhile, the statistical properties of the approximation errors are also utilized to reduce the conservativeness of the filter. The probability distributions of approximation errors are described by sample particles. Coupled with the unknown but bounded noises, the linearized system is associated with two types of uncertainties. To this end, the merging SM and stochastic strategy is adopted to construct a class of confidence state set whose level approaches 1. After that, the optimal filter gain is given to minimize the size of the confidence state set. Finally, three numerical examples are given to demonstrate the effectiveness of the proposed method.

ELECTROMAGNETIC COUPLING OF PIEZOELECTRIC/FERRITE COMPOSITE WITH INITIAL STRESS STATE

A mathematical model of electromagnetic thermoelasticity for an initial-stressed transversal-isotropic composite with piezoelectric magnetostrictive phases has been developed. To solve the related boundary value problem of electromagnetic thermoe-lasticity, the Green function method was used as part of a generalized singular approximation of the statistical mechanics of composites, taking into account the initial stress state of the representative domain of the composite at micro- and macro-levels. The solution of the problem of "effective module" for tensors of effective elastic, piezomechanical, magnetostrictive properties, dielectric permittivity and magnetic permit-tivity, temperature, pyroelectric, pyromagnetic coefficients and (which appeared only at the macro level) electromagnetic and magnetoelectric couplings of a quasi-periodic composite with an initial electromagnetic elastic state was obtained. The solutions for the desired tensors of effective properties of the quasi-periodic composite are presented in the form of analytical formulas of simple linear decompositions by solutions for tensors of effective properties of the periodic structure and statistical mixture, decomposition coefficients are the coefficient of "periodicity" (correlation of quasi-periodic and periodic structures) p and "disordering" 1-p, respectively. Results of calculation of all independent components of tensors of effective coefficients of electromagnetic and magnetoelectric couplings of different structures (periodic, quasi-periodic and statistical mixture) are presented in unidirectional direction of fibrous composite "PZT-4/ferrite" with axisymmet-ric tensor of initial macrostrain of composite. For a quasi-periodic composite (with initial macrostrain), the significantly non-monotonic nature of the dependencies of relative (to values in the absence of an initial stress state) values of effective coefficients of electromagnetic and magnetoelectric couplings from the volume fraction of ferrite fibers was revealed. It was revealed that we have an increase in the absolute values of the electromagnetic and magnetoelectric coupling coefficients of composite at negative values of its initial axisymmetric axial macrostrains. The most significant effect is the initial comprehensive macrostrain in the transversal plane on the coefficients of the transversal magnetoelectric coupling of the composite.

An Energy-Efficient Domain-Specific Reconfigurable Array Processor With Heterogeneous PEs for Wearable Brain–Computer Interface SoCs

Recently, there is increasing demand for energy-efficient signal processing in wearable visual-stimuli-based brain-computer interface (V-BCI) devices. For the better accuracy and the reduced latency of the V-BCI system, the target identification (TI) algorithm that analyzes brain signals is being advanced, and the importance of an energy-efficient accelerating chip that processes various linear algebra operations constituting the TI algorithms is growing. In this paper, we propose a domain-specific reconfigurable array processor (RAP) with a dynamically reconfigurable and scalable array including 5-heterogeneous processing elements (PEs) for the energy-efficient acceleration of basic linear algebra subprograms (BLAS) and matrix decompositions. The system-on-chip (SoC), including the proposed RAP, was fabricated in 130-nm CMOS technology with an area of 16.87-mm2 and measured at 1.0 V 90 MHz. The RAP achieved an information transfer rate (ITR) of 139.9-bits/min and a TI accuracy of 95.4% on a fabricated chip through an optimized TI algorithm and scalable array processing. In addition, the RAP has <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$16.8\times $ </tex-math></inline-formula> higher TI energy efficiency than prior work and achieved an energy efficiency of 2144.2-bits/min/mW for information transfer processing rate with the proposed TI algorithm. The RAP supports a greater variety of linear algebra operations and data sizes with hardware reconfiguration than the prior accelerators.

Physics-constrained deep learning of nonlinear normal modes of spatiotemporal fluid flow dynamics

In this study, we present a physics-constrained deep learning method to discover and visualize from data the invariant nonlinear normal modes (NNMs) which contain the spatiotemporal dynamics of the fluid flow potentially containing strong nonlinearity. Specifically, we develop a NNM-physics-constrained convolutional autoencoder (NNM-CNN-AE) integrated with a multi-temporal-step dynamics prediction block to learn the nonlinear modal transformation, the NNMs containing the spatiotemporal dynamics of the flow, and reduced-order reconstruction and long-time future-state prediction of the flow fields, simultaneously. In test cases, we apply the developed method to analyze different flow regimes past a cylinder, including laminar flows with low Reynolds number in transient and steady states (RD = 100) and high Reynolds number flow (RD = 1000), respectively. The results indicate that the identified NNMs are able to reveal the nonlinear spatiotemporal dynamics of these flows, and the NNMs-based reduced-order modeling consistently achieves better accuracy with orders of magnitudes smaller errors in construction and prediction of the nonlinear velocity and vorticity fields, compared to the linear proper orthogonal decomposition (POD) method and the Koopman-constrained-CNN-AE using the same number or dimension of modes. We perform an analysis of the modal energy distribution of NNMs and find that compared to POD modes, the few fundamental NNMs capture a very high level of total energy of the flow, which is advantageous for reduced-order modeling and representation of the complex flows. Finally, we discuss the potentials and limitations of the presented method.

A closer look at disparities in earnings between white and minoritized dentists.

To examine the factors that account for differences in dentist earnings between White and minoritized dentists. We used data from the American Dental Association's Survey of dental practice, which includes information on 2001-2018 dentist net income, practice ZIP code, patient mix between private and public insurance, and dentist gender, age, and year of dental school graduation. We merged the data on dentist race and ethnicity and school of graduation from the American Dental Association masterfile. Based on practice ZIP code, we also merged the data on local area racial and ethnic composition from the American Community Survey. We used a linear Blinder-Oaxaca decomposition to assess observable characteristics that explain the gap in earnings between White and minoritized dentists. To assess differences in earnings between White and minoritized dentists at different points of the income distribution, we used a re-centered influence function and estimated an unconditional quantile Blinder-Oaxaca decomposition. We extracted data for 22,086 dentists ages 25-85 who worked at least 8 weeks per year and 20 hours per week. Observable characteristics accounted for 58% of the earnings gap between White and Asian dentists, 55% of the gap between White and Hispanic dentists, and 31% of the gap between White and Black dentists. The gap in earnings between White and Asian dentists narrowed at higher quantiles of the income distribution. Compared to other minoritized dentists, Black dentists have the largest earnings disparities relative to White dentists. While the level of the explained component of the disparity for Black dentists is comparable to the explained part of the disparities for other minoritized dentists, the excess percentage of the unexplained component for Black dentists accounts for the additional amount of disparity that Black dentists experienced. Persistent income disparities could discourage minoritized dentists from entering the profession.

The Churches They Are a Changin’: Processes of Change in Worship Services

Abstract One of the most significant recent changes in American religious congregations is the shift toward more enthusiastic worship practices. Despite how clear and significant this trend is, the processes by which worship services have become more enthusiastic remain unclear. I address this question using linear decomposition techniques with data from the National Congregations Study—a nationally representative repeated cross-sectional survey of American religious congregations. Drawing from theories of population ecology, organizational adaptation, and institutions, I identify two distinct processes by which the population of congregations is incorporating more enthusiastic worship practices: (1) new congregations that are disproportionately nondenominational or evangelical are entering the population with highly enthusiastic worship practices, and (2) older congregations across all traditions are changing their worship to incorporate more enthusiastic practices. Because most congregations are older, the second process accounts for most of the change in the population. These findings help advance our understanding of how the population of religious congregations is changing in the 21st century.

Novel uterine contraction signals decomposition for enhanced preterm and birth imminency prediction

• Application of an Intelligent System to tackle a global epidemic as declared by the World Health Organisation • Use of an Intelligent system in Obstetric Medicine • Metaheuristic based signal decomposition for greater prediction accuracy • Explainable Decision Support model Preterm births are one of the key causes of death in children under the age of five: they have financial implications associated with care and cause great psychological distress for the families involved. In this work, we applied a novel signal decomposition approach termed Linear Series Decomposition Learner (LSDL) from electrohysterogram (EHG) and tocodynamometer (Toco) signals to the prediction of a preterm delivery, alongside an associated delivery imminency timeline, using the logistic regression and support vector machine classifiers. The results from the classification exercise showed an equivalent performance for the EHG and Toco signals for the preterm prediction, while in the case of the imminency prediction, the Toco signals provided better results in predicting the delivery imminency period. These results have made apparent that a LSDL decomposed Toco/mechanical signal carries useful information which can be used to predict delivery imminency and supersedes that of an electrophysiological/EHG signal.