Search

Chorioamnionitis is an important cause of preterm delivery. Data on neurodevelopmental outcome in exposed infants are inconsistent due to difficulties in diagnosing intrauterine infection/inflammation and lack of detailed long-term follow-up. We investigate cognitive and motor function in preterm infants at early school age and relate the findings to bacteria in amniotic fluid obtained by amniocentesis (microbial invasion of the amniotic cavity (MIAC)) or placenta findings of histological chorioamnionitis (HCA) or fetal inflammatory response syndrome (FIRS). Sixty-six infants with gestational age

The development of immunoassays enables more sophisticated studies of the associations between protein concentrations and pregnancy outcomes, allowing early biomarker identification that can improve neonatal outcomes. The aim of this study was to explore associations between selected mid-trimester amniotic fluid proteins and (1) overall gestational duration and (2) spontaneous preterm delivery. A prospective cohort study, including women undergoing mid-trimester transabdominal genetic amniocentesis, was performed in Gothenburg, Sweden, 2008–2016 (n = 1072). A panel of 27 proteins related to inflammation was analyzed using Meso-Scale multiplex technology. Concentrations were adjusted for gestational age at sampling, experimental factors, year of sampling, and covariates (maternal age at sampling, parity (nulliparous/multiparous), smoking at first prenatal visit, and in vitro fertilization). Cox regression analysis of the entire cohort was performed to explore possible associations between protein concentrations and gestational duration. This was followed by Cox regression analysis censored at 259 days or longer, to investigate whether associations were detectable in women with spontaneous preterm delivery (n = 47). Finally, linear regression models were performed to analyze associations between protein concentrations and gestational duration in women with spontaneous onset of labor at term (n = 784). HMG-1, IGFBP-1, IL-18, MIP-1α, MIP-1β, S100A8, and thrombospondin-1 were significantly associated with gestational duration at term, but not preterm. Increased concentrations of thrombospondin-1, MIP-1β, and S100A8, respectively, were significantly associated with decreased gestational duration after the Holm-Bonferroni correction in women with spontaneous onset of labor at term. This adds to the concept of a pregnancy clock, where our findings suggest that such a clock is also reflected in the amniotic fluid at early mid-trimester, but further research is needed to confirm this.

Parametric dictionaries can increase the ability of sparse representations to meaningfully capture and interpret the underlying signal information, such as encountered in biomedical problems. Given a mapping function from the atom parameter space to the actual atoms, we propose a sparse Bayesian framework for learning the atom parameters, because of its ability to provide full posterior estimates, take uncertainty into account and generalize on unseen data. Inference is performed with Markov Chain Monte Carlo, that uses block sampling to generate the variables of the Bayesian problem. Since the parameterization of dictionary atoms results in posteriors that cannot be analytically computed, we use a Metropolis–Hastings–within-Gibbs framework, according to which variables with closed-form posteriors are generated with the Gibbs sampler, while the remaining ones with the Metropolis Hastings from appropriate candidate-generating densities. We further show that the corresponding Markov Chain is uniformly ergodic ensuring its convergence to a stationary distribution independently of the initial state. Results on synthetic data and real biomedical signals indicate that our approach offers advantages in terms of signal reconstruction compared to previously proposed Steepest Descent and Equiangular Tight Frame methods. This paper demonstrates the ability of Bayesian learning to generate parametric dictionaries that can reliably represent the exemplar data and provides the foundation towards inferring the entire variable set of the sparse approximation problem for signal denoising, adaptation, and other applications.

Objective: The aim of this study was to explore inflammatory response and identify early potential biomarkers in mid-trimester amniotic fluid associated with subsequent spontaneous preterm delivery (PTD). Methods: A cohort study was performed at Sahlgrenska University Hospital/Ostra, Gothenburg, Sweden, between 2008 and 2010. Amniotic fluid was collected from consecutive women undergoing mid-trimester transabdominal genetic amniocentesis at 14–19 gestational weeks. Clinical data and delivery outcome variables were obtained from medical records. The analysis included 19 women with spontaneous PTD and 118 women who delivered at term. A panel of 26 candidate proteins was analyzed using Luminex xMAP technology. Candidate protein concentrations were analyzed with ANCOVA and adjusted for plate effects. Results: The median gestational age at delivery was 35 + 3 weeks in women with spontaneous PTD and 40 + 0 weeks in women who delivered at term. Nominally significantly lower amniotic fluid levels of adiponectin (PTD: median 130,695 pg/mL (IQR 71,852–199,414) vs term: median 185,329 pg/mL (IQR (135,815–290,532)), granulocyte-macrophage colony stimulating factor (PTD: median 137 pg/mL (IQR 74–156) vs term: median 176 pg/mL (IQR 111–262)), and macrophage migration inhibitory factor (PTD: median 3025 pg/mL (IQR 1885–3891) vs term: median 3400 pg/mL (IQR 2181–5231)) were observed in the spontaneous PTD group, compared with the term delivery group, after adjusting for plate effects. No significant differences remained after Bonferroni correction for multiple comparisons. Conclusions: Our results are important in the process of determining the etiology behind spontaneous PTD but due to the non-significance after Bonferroni correction, the results should be interpreted with caution. Further analyses of larger sample size will be required to determine whether these results are cogent and to examine whether microbial invasion of the amniotic cavity or intra-amniotic inflammation occurs in asymptomatic women in the mid-trimester with subsequent spontaneous PTD.

We describe the methodology for the collection and annotation of a large corpus of emotional speech data through crowdsourcing. The corpus offers 187 hours of data from 2,965 subjects. Data includes non-emotional recordings from each subject as well as recordings for five emotions: angry, happy-low-arousal, happy-high-arousal, neutral, and sad. The data consist of spontaneous speech elicited from subjects via a web-based tool. Subjects used their own personal recording equipment, resulting in a data set that contains variation in room acoustics, microphone, etc. This offers the advantage of matching the type of variation one would expect to see when exposing speech technology in the wild in a web-based environment. The annotation scheme covers the quality of emotion expressed through the tone of voice and what was said, along with common audio-quality issues. We discuss lessons learned in the process of the creation of this corpus.

Biometric sensors and portable devices are being increasingly embedded into our everyday life, creating the need for robust physiological models that efficiently represent, analyze, and interpret the acquired signals. We propose a knowledge-driven method to represent electrodermal activity (EDA), a psychophysiological signal linked to stress, affect, and cognitive processing. We build EDA-specific dictionaries that accurately model both the slow varying tonic part and the signal fluctuations, called skin conductance responses (SCR), and use greedy sparse representation techniques to decompose the signal into a small number of atoms from the dictionary. Quantitative evaluation of our method considers signal reconstruction, compression rate, and information retrieval measures, that capture the ability of the model to incorporate the main signal characteristics, such as SCR occurrences. Compared to previous studies fitting a predetermined structure to the signal, results indicate that our approach provides benefits across all aforementioned criteria. This paper demonstrates the ability of appropriate dictionaries along with sparse decomposition methods to reliably represent EDA signals and provides a foundation for automatic measurement of SCR characteristics and the extraction of meaningful EDA features.

Wearable technology permeates every aspect of our daily life increasing the need of reliable and interpretable models for processing the large amount of biomedical data. We propose the EDA-Gram, a multidimensional fingerprint of the electrodermal activity (EDA) signal, inspired by the widely-used notion of spectrogram. The EDA-Gram is based on the sparse decomposition of EDA from a knowledge-driven set of dictionary atoms. The time axis reflects the analysis frames, the spectral dimension depicts the width of selected dictionary atoms, while intensity values are computed from the atom coefficients. In this way, EDA-Gram incorporates the amplitude and shape of Skin Conductance Responses (SCR), which comprise an essential part of the signal. EDA-Gram is further used as a foundation for signal-specific feature design. Our results indicate that the proposed representation can accentuate fine-grain signal fluctuations, which might not always be apparent through simple visual inspection. Statistical analysis and classification/regression experiments further suggest that the derived features can differentiate between multiple arousal levels and stress-eliciting environments for two datasets.

Growth in voice-based applications and personalized systems has led to increasing demand for speech- analytics technologies that estimate the state of a speaker from speech. Such systems support a wide range of applications, from more traditional call-center monitoring, to health monitoring, to human-robot interactions, and more. To work seamlessly in real-world contexts, such systems must meet certain requirements, including for speed, customizability, ease of use, robustness, and live integration of both acoustic and lexical cues. This demo introduces SenSay AnalyticsTM, a platform that performs real-time speaker-state classification from spoken audio. SenSay is easily configured and is customizable to new domains, while its underlying architecture offers extensibility and scalability.

Objective: To determine the impact of histological chorioamnionitis (HCA) and funisitis on neonatal outcome in preterm prelabor rupture of membranes (PPROM) pregnancies.Methods: Women with PPROM between 24 + 0 to 36 + 6 weeks of gestation, admitted to the Department of Obstetrics and Gynecology at the University Hospital Hradec Kralove in the Czech Republic, between July 2008 and October 2010, were enrolled in the study (n = 231).Results: The incidence of early-onset sepsis (EOS) differed significantly in neonates born to women with and without HCA, after adjustment for gestational age (11% versus 1%, p = 0.011). The incidence of EOS in neonates was also significantly different, after adjustment for gestational age, in cases with and without funisitis (18% versus 4%, p = 0.002). The same was also found for retinopathy of prematurity (ROP) cases with and without funisitis (23% versus 4%, p = 0.014), after adjustment for gestational age.Conclusions: HCA and funisitis increase the risk of adverse per...

BACKGROUND: Prediction of neonatal respiratory morbidity may be useful to plan delivery in complicated pregnancies. The limited predictive performance of the current diagnostic tests together with the risks of an invasive procedure restricts the use of fetal lung maturity assessment. OBJECTIVE: The objective of the study was to evaluate the performance of quantitative ultrasound texture analysis of the fetal lung (quantusFLM) to predict neonatal respiratory morbidity in preterm and early-term (

In this paper, we investigate the task of translating spontaneous speech transcriptions by employing aligned movie subtitles in training a statistical machine translator (SMT). In contrast to the lexical-based dynamic time warping (DTW) approaches to bilingual subtitle alignment, we align subtitle documents using time-stamps. We show that subtitle time-stamps in two languages are often approximately linearly related, which can be exploited for extracting high-quality bilingual subtitle pairs. On a small tagged data-set, we achieve a performance improvement of 0.21 F-score points compared to traditional DTW alignment approach and 0.39 F-score points compared to a simple line-fitting approach. In addition, we achieve a performance gain of 4.88 BLEU score points in spontaneous speech translation experiments using the aligned subtitle data obtained by the proposed alignment approach compared to that obtained by the DTW based alignment approach demonstrating the merit of the time-stamps based subtitle alignment scheme.

Objective. To evaluate the maternal inflammatory response to microbial invasion of the amniotic cavity (MIAC) in women with preterm labor and preterm prelabor rupture of membranes using selected proteins in the maternal serum. Design. A prospective cohort study. Setting. Labor ward from Salgrenska University Hospital. The evaluation of the maternal inflammatory response in the presence of MIAC in preterm labor and preterm prelabor rupture of membranes. Population. One hundred and sixteen women with preterm labor and 73 women with preterm prelabor rupture of membranes between the gestational ages of 22+0 and 33+6 weeks. Methods. Twenty-seven maternal serum proteins were assayed by a multiple immunoassay. Main outcome measures. The maternal serum inflammatory response was evaluated according to the presence of MIAC. Data were stratified by gestational age. Results. There were few differences in the maternal serum protein levels when MIAC was present in both preterm labor and preterm prelabor rupture of membranes. In preterm prelabor rupture of membranes, higher levels of interleukin-18 (median 654 vs. 361 pg/mL, p= 0.003) and lower levels of interleukin-1β (9.5 vs. 19.9 pg/mL, p= 0.008) and monocyte chemotactic protein-1 (139.1 vs. 212.6 pg/mL, p= 0.039) were observed in women with MIAC. Interleukin-6 (20.8 vs. 13.9 pg/mL, p= 0.019) was the only biomarker that increased significantly in preterm labor complicated with MIAC. All of the differences between preterm labor and preterm prelabor rupture of membranes were observed at less than 32+0 weeks of gestation. Conclusions. A weak maternal inflammatory response in the serum was observed in women with MIAC.

Please cite this paper as: Tsiartas P, Holst R, Wennerholm U, Hagberg H, Hougaard D, Skogstrand K, Pearce B, Thorsen P, Kacerovsky M, Jacobsson B. Prediction of spontaneous preterm delivery in women with threatened preterm labour: a prospective cohort study of multiple proteins in maternal serum. BJOG 2012;119:866–873. Objective To analyse whether specific proteins in maternal serum and cervical length, alone or in combination, can predict the likelihood that women with intact membranes with threatened preterm labour will deliver spontaneously within 7 days of sampling. Design Cohort study. Setting Sahlgrenska University Hospital, Gothenburg, Sweden. Population Women at between 22 and 33 weeks of gestation with threatened preterm labour (n = 142) admitted to the Sahlgrenska University Hospital, Gothenburg, Sweden, in 1995–2005. Methods Maternal serum was tested for 27 proteins using multiplex xMAP technology. Individual levels of each protein were compared, and calculations were performed to investigate potential associations between different proteins, cervical length and spontaneous preterm delivery. Receiver operating characteristic curves were used to find the best cut-off values for continuous variables in relation to spontaneous preterm delivery within 7 days of sampling. Prediction models were created based on a stepwise logistic regression using binary variables. Main outcome measure Spontaneous preterm delivery within 7 days. Results In order to determine the best prediction model, we analysed models of serum proteins alone, cervical length alone, and the combination of serum proteins and cervical length. We found one multivariable combined model through the data analysis that more accurately predicted spontaneous preterm delivery within 7 days. This model was based on serum interleukin-10 (IL-10) levels, serum RANTES levels and cervical length (sensitivity 74%, specificity 87%, positive predictive value 76%, negative predictive value 86%, likelihood ratio 5.8 and area under the curve 0.88). Conclusions A combination of maternal serum proteins and cervical length constituted the best prediction model, and would help determine whether women with threatened preterm labour are likely to deliver within 7 days of measurement.

The co-variation degree between individuals in their physiological signals can reveal insights about the quality of their interaction as well as their personal characteristics. In an effort to capture the amount of synchrony between Electrodermal Activity (EDA) streams occurring in parallel during dyadic interactions, we propose Sparse EDA Synchrony Measure (SESM), an index derived from the joint sparse representation of EDA ensembles. Sparse decomposition is performed using Simultaneous Orthogonal Matching Pursuit (SOMP) from a knowledge-driven dictionary of tonic and phasic atoms, capturing the slow-varying trends and high-frequency signal fluctuations, respectively. At each iteration the atom having the maximum average correlation with the residuals is selected. We compute SESM as the negative natural logarithm of the joint reconstruction error and evaluate it with data from interactions of married and young dating couples participating in tasks of varying emotional intensity. Through statistical analysis and multiple linear regression experiments, our results indicate that SESM depicts significant differences across tasks in both datasets considered and can be associated to individuals' attachment-related characteristics.

We propose the energy-constrained minimum-variance response (ECMVR) filter to perform robust spectral estimation of vowels. We modify the distortionless constraint of the minimum-variance distortionless response (MVDR) filter and add an energy constraint to its formulation to mitigate the influence of noise on the speech spectrum. We test our ECMVR filter on a vowel classification task with different background noises at various SNR levels. Results show that vowels are classified more accurately in certain noises using MFCC and PLP features extracted from the ECMVR spectrum compared to using features extracted from the FFT and MVDR spectra.

This paper introduces a supervised statistical framework for estimating the signal-to-noise (SNR) ratio of speech signals. Information on how noise corrupts a signal can help us compensate for its effects, especially in real life applications where the usual assumption of white Gaussian noise does not hold and speech boundaries in the signal are not known. We use features from which we can detect speech regions in a signal, without using Voice Activity Detection, and estimate the energies of those regions. Then we use these features to train ordinary least squares regression models for various noise types. We compare this supervised method with state-of-the-art SNR estimation algorithms and show its superior performance with respect to the tested noise types.

In this paper, we propose robust features for the problem of voice activity detection (VAD). In particular, we extend the long term signal variability (LTSV) feature to accommodate multiple spectral bands. The motivation of the multi-band approach stems from the non-uniform frequency scale of speech phonemes and noise characteristics. Our analysis shows that the multi-band approach offers advantages over the single band LTSV for voice activity detection. In terms of classification accuracy, we show 0.3%-61.2% relative improvement over the best accuracy of the baselines considered for 7 out 8 different noisy channels. Experimental results, and error analysis, are reported on the DARPA RATS corpora of noisy speech. Index Terms: noisy speech data, voice activity detection, robust feature extraction

Despite the increasing interest in Speech-to-speech (S2S) translation, research and development has focused almost exclusively on the lexical aspects of translation. The importance of transferring prosodic and other paralinguistic information through S2S devices and evaluating its impact on the translation quality are yet to be well established. The novelty in this work is a large scale human evaluation study to test the hypothesis that cross-lingual prosodic emphasis transfer is directly related to the perceived quality of speech translation. This hypothesis is validated at the 0.53-0.54 correlation level on the data sets considered with results significant at p-value=0.01. The second contribution of this work is an evaluation methodology based on crowd sourcing using English-Spanish language bilingual data from two distinct domains and evaluated with over 200 bilingual speakers. We also present lessons learned on this type of S2S subjective experiments when using crowd sourcing.

This paper presents a simplified and supervised i-vector modeling framework that is applied in the task of robust and efficient speaker verification (SRE). First, by concatenating the mean supervector and the i-vector factor loading matrix with respectively the label vector and the linear classifier matrix, the traditional i-vectors are then extended to label-regularized supervised i-vectors. These supervised i-vectors are optimized to not only reconstruct the mean supervectors well but also minimize the mean squared error between the original and the reconstructed label vectors, such that they become more discriminative. Second, factor analysis (FA) can be performed on the pre-normalized centered GMM first order statistics supervector to ensure that the Gaussian statistics sub-vector of each Gaussian component is treated equally in the FA, which reduces the computational cost significantly. Experimental results are reported on the female part of the NIST SRE 2010 task with common condition 5. The proposed supervised i-vector approach outperforms the i-vector baseline by relatively 12% and 7% in terms of equal error rate (EER) and norm old minDCF values, respectively.

Extraction of bilingual audio and text data is crucial for designing Speech to Speech (S2S) systems. In this work, we propose an automatic method to segment multilingual audio streams from movies. In addition, the audio streams are aligned with the corresponding subtitles. We found that the proposed method gives 89% perfectly segmented bilingual audio and 6% partially segmented bilingual audio. In addition, the mapping of the audio to the corresponding subtitles has accuracy 91%.

In this paper, we consider the estimation of topic specific Language Models (LM) by exploiting documents from the World Wide Web (WWW). We focus on the quality of the generated queries and propose a novel query generation method. In contrast to the n-gram based queries used in past works, our approach relies on utterances as queries candidates. The proposed approach does not rely on any language specific information other than the initial in-domain training text. We have conducted experiments with Web texts of size 0–150 million words, and we have shown that despite not using any language specific information, the proposed approach results in up to 1.1% absolute Word Error Rate (WER) improvement as compared to keyword-based approaches. The proposed approach reduces the WER by 6.3% absolute in our experiments, compared to an in-domain LM without considering any Web data.

Application of an external electric field across thin polymeric films is known to induce the formation of arrays of cylindrical structures commonly referred to as pillars. Low viscosity photocurable materials have been shown to possess significant advantages over their polymeric counterparts both in terms of a significant reduction in the time of formation, as well as in the elimination of additional processing steps, such as heating and cooling cycles. With proper design, pillar arrays with predetermined diameter, period, location and orientation can theoretically be formed over an entire substrate surface, thus enabling their use in applications such as optics-less patterning, micro-electro-mechanical systems, and micro fluidic devices. In this paper, we present our recent work in using a low viscosity thiolene system to produce pillar arrays and attempts at inducing long range order in the resulting pillar arrays by (1) varying the nature of the external electric field in the form of a sinusoidal AC voltage (instead of the conventional DC voltage) and (2) the use of ultrasonic piezo transducers to generate regular waves in the film with simultaneous application of a potential across the film sample. Our initial results show that although both the AC voltage and the ultrasonic piezo methods are successful in producing pillar arrays, their ordering is short-ranged and does not extend throughout the entire sample area.

Optical lithography with special thick film DNQ-novolac photoresists have been practiced for many years to fabricate microstructures that require feature heights ranging from several to hundreds of microns such as thin film magnetic heads. It is common in these thick film photoresist systems to observe interesting non-uniform profiles with narrow regions near the top surface of the film that transition into broader and more concave shapes near the bottom of the resist profile. A number of explanations have been proposed for these various observations including the formation of `dry skins' at the resist surface and the presence of solvent gradients in the film which serve to modify the local development rate of the photoresist. There have been few detailed experimental studies of the development behavior of thick films resists. This has been due to part to the difficulty in studying these films with conventional dissolution rate monitors (DRMs). In general, this lack of experimental data along with other factors has made simulation and modeling of thick film resist performance difficult. As applications such as thin film head manufacturing drive to smaller features with higher aspect ratios, the need for accurate thick film simulation capability continues to grow. A new multi-wavelength DRM tool has been constructed and used in conjunction with a resist bleaching tool and rigorous parameter extraction techniques to establish exposure and development parameters for two thick film resists, AZTM 4330-RS and AZTM 9200. Simulations based on these parameters show good agreement to resist profiles for these two resists.

Accurate photoresist modeling parameters are required for correct lithographic simulations. In particular, three sets of data are required to model a typical non-chemically amplified resist: the refractive index as a function of wavelength and exposure, the exposure or ABC parameters (the Dill parameters), and the development parameters. This work focuses on an improved technique for the accurate extraction of development rate parameters for non-chemically amplified resists. Traditionally, the refractive index for photoresists is considered constant in simulations and other work. However, the refractive index of a photoresist varies as the chemical composition of the photoresist changes during exposure. This work presents a rigorous analysis technique for extracting development parameters from resist development rate data which accounts for this refractive index change. Development parameter measurements for a commercial resist will be discussed. Comparisons of the various development rate models will be made using this experimental data.

In the past, resist parameters (exposure and development parameters) were typically only available for a single set of processing conditions. Therefore, it has been impossible to explore the effect of processing conditions on resist performance using simulation. In this work, a statistical experimental design and response surface analysis technique was used in conjunction with our improved parameter extraction techniques to investigate the effect of processing conditions on the exposure and development parameters for a commercial i- line resist. The effect of soft bake time and temperature on the exposure parameters and the effect of soft bake temperature, soft bake time, post-exposure bake temperature, and post-exposure bake time on development parameters is discussed. Using this information, it is possible for the first time to consider optimizing resist processing conditions using lithographic simulations.

Intrapartum application of the continuous glucose monitoring system in pregnancies complicated with diabetes: A review and feasibility study

Objective The objective of the study was to determine whether the bacterial load of genital mycoplasmas and gestation age are related to intraamniotic inflammatory response using the amniotic fluid levels of 18 inflammatory mediators. Study Design A prospective study of 145 women with singleton pregnancies complicated by preterm prelabor rupture of membranes between 24 0/7 and 36 6/7 weeks was conducted. Amniotic fluid was obtained from all women by transabdominal amniocentesis. The amounts of genital mycoplasma deoxyribonucleic acid were determined using the threshold cycle value and relative and absolute quantification techniques. A panel of multiple proteins was analyzed simultaneously using multiplex technology. Results Twenty-four women with the presence of genital mycoplasmas in amniotic fluid were included in the final analyses. The concentrations of 9 of the 18 evaluated proteins in the amniotic fluid correlated with bacterial load of genital mycoplasmas independent of the quantification technique used. Conclusion The intensity of intraamniotic inflammatory response to genital mycoplasmas decreased with gestational age.

delivery in women with preterm labor: analysis of multiple proteins in maternal serum Panagiotis Tsiartas, Rose-Marie Holst, Ulla-Britt Wennerholm, Henrik Hagberg, David Hougaard, Kristin Skogstrand, Brad Pearce, Marian Kacerovsky, Bo Jacobsson Papageorgiou University Hospital of Thessaloniki, Obstetrics and Gynecology, Thessaloniki, Greece, Sahlgrenska University Hospital, Department of Obstetrics and Gynecology, Gothenburg, Sweden, Sahlgrenska University Hospital, East, Obstetrics and Gynecology, Gothenburg, Sweden, Queen Charlottes and Chelsea Hospital, Institute of Reproductive and Developmental Biology, London, United Kingdom, Statens Serum Institut, Department of Clinical Biochemistry and Immunology, Copenhagen, Denmark, Statens Serum Institut, 3Department of Clinical Biochemistry and Immunology, Copenhagen, Denmark, Rollins School of Public Health, Epidemiology, Atlanta, GA, Univerzita Karlova Praha, Porodnicka a gynekologicka klinika FN Hradec Kralova, Praha, Czech Republic OBJECTIVE: To analyze whether specific proteins in maternal serum and the cervical length, alone or in combination, can identify the likelihood that women in preterm labor with intact membranes will deliver spontaneously within 7 days of sampling. STUDY DESIGN: We used in our study two independent cohorts of women, an exploratory cohort from 1995-2005 and a confirmatory cohort from 2008-2010 admitted to the Sahlgrenska University Hospital, Gothenburg, Sweden in preterm labor. In the exploratory cohort of 142 women in preterm labor between 22 and 33 weeks of gestation, maternal serum was collected and tested for twenty-seven proteins using multiplex xMAP technology. The women in the study provided serum samples at admission to the hospital for xMAP analysis. Individual levels of each protein were compared and calculations performed to find associations among different proteins, cervical length, and spontaneous preterm delivery within 7 days of sampling. Prediction models were created based on stepwise logistic regression. We have in a second stage applied our multivariable model in order to confirm its predictive capacity to an independent confirmatory cohort of 42 women between 23 and 33 weeks of gestation, who were admitted to the hospital in preterm labor and provided serum samples for xMAP analysis. RESULTS: We found one multivariable model through the data analysis of the exploratory cohort of women that predicts spontaneous preterm delivery within 7 days. This model was based on serum IL-10, RANTES and cervical length (sensitivity 73.8%, specificity 87.3%, positive predictive value 75.6%, negative predictive value 86.3%, likelihood ratio 5.8 and area under the curve 0.89). From the application of the multivariable model to the confirmatory cohort we found that the model had a sensitivity of 66.6%, specificity 66.6%, positive predictive value 35.3%, negative predictive value 88%, likelihood ratio 2 and area under the curve 0.66. CONCLUSION: A combination of maternal serum proteins and cervical length might help to determine whether women with preterm labor will deliver within 7 days.