Detection of Maternal and Fetal Stress from ECG with Self-supervised Representation Learning

Published on Nov 3, 2020in arXiv: Quantitative Methods
Pritam Sarkar5
Estimated H-index: 5
,
Silvia M. Lobmaier12
Estimated H-index: 12
+ 5 AuthorsAli Etemad8
Estimated H-index: 8
Sources
Abstract
In a pregnant mother and her fetus, chronic prenatal stress results in entrainment of the fetal heartbeat by the maternal heartbeat, quantified by the fetal stress index (FSI). Deep learning (DL) is capable of pattern detection in complex medical data with high accuracy in noisy real-life environments, but little is known about the utility of DL in non-invasive biometrics during pregnancy. We hypothesized that a recently established self-supervised learning (SSL) approach that provides emotional recognition from ECG will identify chronically stressed mother-fetus dyads from the raw maternal abdominal electrocardiograms (aECG), containing fetal and maternal ECG. Chronically stressed mothers and controls matched on enrolment at 32 weeks of gestation were studied. We validated the chronic stress exposure by psychological inventory, maternal hair cortisol, and FSI. We tested two variants of SSL architecture, one trained on the generic ECG features for emotional recognition obtained from public datasets and another transfer-learned on a subset of our data. Our DL models predict excellently the chronic stress exposure group (AUROC 0.982+/-0.002), the individual psychological stress score (R2 0.943+/-0.009), and FSI at 34 weeks of gestation (R2 0.946+/-0.013) as well as the maternal hair cortisol at birth reflecting chronic stress exposure (R2 0.931+/-0.006). The best performance was achieved with the DL model trained on the public dataset and using maternal ECG alone. The present DL approach provides a novel source of physiological insights into complex multi-modal relationships between different regulatory systems exposed to chronic stress. The developed DL model can be deployed in low-cost regular ECG biosensors as a simple ubiquitous early stress detection and exposure tool during pregnancy. This discovery should enable early behavioral interventions.
References20
Newest
#1Pritam Sarkar (Queen's University)H-Index: 5
#2Ali Etemad (Queen's University)H-Index: 5
Last. Ali Etemad (Queen's University)H-Index: 8
view all 2 authors...
We exploit a self-supervised deep multi-task learning framework for electrocardiogram (ECG) -based emotion recognition. The proposed solution consists of two stages of learning a) learning ECG representations and b) learning to classify emotions. ECG representations are learned by a signal transformation recognition network. The network learns high-level abstract representations from unlabeled ECG data. Six different signal transformations are applied to the ECG signals, and transformation recog...
17 CitationsSource
#1Paula Desplats (UCSD: University of California, San Diego)H-Index: 33
#2Ashley M. Gutierrez (UCSD: University of California, San Diego)H-Index: 3
Last. Martin G. Frasch (UW: University of Washington)H-Index: 20
view all 4 authors...
Abstract We review evidence supporting the role of early life programming in the susceptibility for adult neurodegenerative diseases while highlighting questions and proposing avenues for future research to advance our understanding of this fundamental process. The key elements of this phenomenon are chronic stress, neuroinflammation triggering microglial polarization, microglial memory and their connection to neurodegeneration. We review the mediating mechanisms which may function as early biom...
17 CitationsSource
#1Martin G. Frasch (UW: University of Washington)H-Index: 20
#2Silvia M. Lobmaier (TUM: Technische Universität München)H-Index: 12
Last. Marta C. Antonelli (UBA: University of Buenos Aires)H-Index: 24
view all 14 authors...
Abstract Prenatal stress (PS) impacts early postnatal behavioural and cognitive development. This process of ‘fetal programming’ is mediated by the effects of the prenatal experience on the developing hypothalamic–pituitary–adrenal (HPA) axis and autonomic nervous system (ANS). We derive a multi-scale multi-species approach to devising preclinical and clinical studies to identify early non-invasively available pre- and postnatal biomarkers of PS. The multiple scales include brain epigenome, meta...
18 CitationsSource
#1Hyoyoung Jeong (NU: Northwestern University)H-Index: 9
#2John A. Rogers (NU: Northwestern University)H-Index: 182
Last. Shuai Xu (NU: Northwestern University)H-Index: 22
view all 3 authors...
As of June 20th, 2020, the Center for Disease Control’s tabulations show more than 2.2 million recorded cases of COVID-19 and nearly 120,000 in deaths in the U.S. ( 1 ) Infected patients present with a wide range of symptoms, from completely asymptomatic to rapidly progressive pneumonia leading to death. Rigorous and widespread testing remains a critical component of strategies for containing this pandemic. The limited availability of molecular diagnostics constrains the use of these technologie...
44 CitationsSource
May 4, 2020 in ICASSP (International Conference on Acoustics, Speech, and Signal Processing)
#1Pritam Sarkar (Queen's University)H-Index: 5
#2Ali Etemad (Queen's University)H-Index: 8
We present an electrocardiogram (ECG) -based emotion recognition system using self-supervised learning. Our proposed architecture consists of two main networks, a signal transformation recognition network and an emotion recognition network. First, unlabelled data are used to successfully train the former network to detect specific pre-determined signal transformations in the self-supervised learning step. Next, the weights of the convolutional layers of this network are transferred to the emotio...
17 CitationsSource
#1Silvia M. Lobmaier (TUM: Technische Universität München)H-Index: 12
#2Alexander Müller (TUM: Technische Universität München)H-Index: 7
Last. Marta C. Antonelli (UBA: University of Buenos Aires)H-Index: 24
view all 13 authors...
Purpose Prenatal stress (PS) during pregnancy affects in utero- and postnatal child brain-development. Key systems affected are the hypothalamic–pituitary–adrenal axis and the autonomic nervous system (ANS). Maternal- and fetal ANS activity can be gauged non-invasively from transabdominal electrocardiogram (taECG). We propose a novel approach to assess couplings between maternal (mHR) and fetal heart rate (fHR) as a new biomarker for PS based on bivariate phase-rectified signal averaging (BPRSA)...
12 CitationsSource
#1Kyle Ross (Queen's University)H-Index: 2
#2Pritam Sarkar (Queen's University)H-Index: 5
Last. Ali Etemad (Queen's University)H-Index: 8
view all 8 authors...
Simulation-based training has been proven to be a highly effective pedagogical strategy. However, misalignment between the participant’s level of expertise and the difficulty of the simulation has been shown to have significant negative impact on learning outcomes. To ensure that learning outcomes are achieved, we propose a novel framework for adaptive simulation with the goal of identifying the level of expertise of the learner, and dynamically modulating the simulation complexity to match the ...
11 CitationsSource
#1Pritam SarkarH-Index: 5
#2Kyle RossH-Index: 2
Last. Ali EtemadH-Index: 8
view all 6 authors...
Simulations are a pedagogical means of enabling a risk-free way for healthcare practitioners to learn, maintain, or enhance their knowledge and skills. Such simulations should provide an optimum amount of cognitive load to the learner and be tailored to their levels of expertise. However, most current simulations are a one-type-fits-all tool used to train different learners regardless of their existing skills, expertise, and ability to handle cognitive load. To address this problem, we propose a...
7 CitationsSource
#1Diego Gonzalez (UBA: University of Buenos Aires)H-Index: 3
#2Dario Jacobsen (UBA: University of Buenos Aires)H-Index: 3
Last. Bibiana Fabre (UBA: University of Buenos Aires)H-Index: 9
view all 12 authors...
We present the development of the first procedure for hair cortisol measurement through an automated method. Hair samples were obtained from 286 individuals. After cortisol extraction, samples were measured in a Siemens Immulite 2000 (Gwynedd, UK) automated chemoluminiscent immunoassay analyzer. Normal reference values were obtained from hair cortisol levels measured in 213 healthy individuals with low levels of stress. Hair cortisol concentration median was 55 pg/mg hair (2.5–97.5 percentile (4...
11 CitationsSource
#1Philip Schmidt (University of Siegen)H-Index: 4
#2Attila Reiss (Bosch)H-Index: 16
Last. Kristof Van Laerhoven (University of Siegen)H-Index: 25
view all 5 authors...
Affect recognition aims to detect a person's affective state based on observables, with the goal to e.g. improve human-computer interaction. Long-term stress is known to have severe implications on wellbeing, which call for continuous and automated stress monitoring systems. However, the affective computing community lacks commonly used standard datasets for wearable stress detection which a) provide multimodal high-quality data, and b) include multiple affective states. Therefore, we introduce ...
129 CitationsSource
Cited By1
Newest
Facial expression recognition (FER) has emerged as an important component of human-computer interaction systems. Despite recent advancements in FER, performance often drops significantly for non-frontal facial images. We propose Contrastive Learning of Multi-view facial Expressions (CL-MEx) to exploit facial images captured simultaneously from different angles towards FER. CL-MEx is a two-step training framework. In the first step, an encoder network is pre-trained with the proposed self-supervi...
Source
We propose a self-supervised contrastive learning approach for facial expression recognition (FER) in videos. We propose a novel temporal sampling-based augmentation scheme to be utilized in addition to standard spatial augmentations used for contrastive learning. Our proposed temporal augmentation scheme randomly picks from one of three temporal sampling techniques: (1) pure random sampling, (2) uniform sampling, and (3) sequential sampling. This is followed by a combination of up to three stan...
1 Citations