This paper presents a deep learning-based approach for reliable fetal QRS detection in abdominal ECG recordings. Fetal electrocardiography (fECG) enables non-invasive monitoring of fetal heart activity using surface electrodes on the mother’s abdomen. Identifying fetal QRS complexes is crucial for heart rate monitoring but challenging due to their low amplitude relative to maternal components and noise. Limited labeled datasets further hinder the development of advanced detection methods. To address this, we extended the FECGSYN toolbox to generate semi-simulated abdominal fECG datasets. Simulated signals were combined with real ECG noise, and additional vectorcardiograms were incorporated to increase variability. This process produced 1,200 records suitable for training deep learning models. As proof of concept, a 1D U-Net was trained to segment fetal QRS regions, achieving a 98

Objective Sub-optimal uterine-placental perfusion and fetal nutrition can lead to intrauterine growth restriction (IUGR), also called fetal growth restriction (FGR). Antenatal cardiotocography (CTG) can aid in the early detection of IUGR. Reliably diagnosing IUGR before delivery remains challenging, and deep learning (DL) techniques offer potential solutions. This paper describes the development of a DL approach to predict an IUGR condition at birth by using CTG signals collected during antenatal monitoring. Materials and methods Our method is encapsulated in the concept of a two-step training process of a ResNet architecture. The primary focus is on the minimization of data loss, which motivates the division into “presumed” and “confirmed” datasets, which is employed to distinguish based on the presence of information at birth. The method involves fine-tuning: the initial training utilizes “presumed” data to train the network, and the subsequent training employs data representing certain knowledge to refine its performance. Results The DL model reaches a balanced accuracy of 80% on a hold-out test set of confirmed cases, which is better than what obtained by using standard clinical guidelines. Discussion The results of our work are compared to the results of similar papers dealing with the prediction of IUGR condition at birth and in general with the prediction of fetal pathological conditions. Our final results are obtained using a very large dataset compared to other papers reported in the literature. Conclusion The inclusion of DL methods on CTG signals may complement imaging technologies and improve the early detection of IUGR.

The secondary use of health data represents a great opportunity to advance pathophysiological knowledge and improve patients’ care. However, the absence of standard data formats and information structuring schemas severely hinders this potential, preventing the efficient sharing of data collected in different hospitals and affecting the quality of multicentric studies. The 10-year Health Big Data (HBD) project aims to address these issues to foster the collaboration of 51 Italian research hospitals (IRCCSs). To address the seven main challenges identified for health data sharing, seven Working Groups (WGs) were created, with the WG2 being responsible for the definition of standardization and harmonization pipelines for signals, bioimages, and omics data. The present paper focuses on two ongoing works of the WG2, namely the implementation of a pipeline to extract and map information from electrocardiographic (ECG) signals into the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) and the development of a harmonization pipeline to reduce the center effect in multicentric Magnetic Resonance Imaging (MRI) studies. We show interesting results and insights concerning the implementation of both pipelines. Besides, we highlight the main difficulties we encountered on our path toward health data sharing and suggest possible solutions.

Computerized Cardiotocography (cCTG) facilitates a thorough and objective examination of the Fetal Heart Rate (FHR), providing valuable insights into the fetal condition and its well-being. A crucial aspect within this context pertains to the automatic identification of periods of fetal activity and quiescence, which are associated with different FHR patterns. The accurate discrimination of these patterns holds the potential to improve the interpretability and diagnostic capabilities of FHR quantitative analysis. Indeed, disruptions in the cycling between active and quiet periods are associated with the development of pathological conditions. This study introduces a deep learning based methodology for the identification of fetal behavioral heart rate patterns. Specifically, the implemented deep neural network (DNN) adopts a 1D encoder-decoder architecture, which is trained to recognize and automatically segment the FHR recordings into active and quiet periods. The proposed framework includes a semi-supervised training process, based on two steps: a) DNN pre-training based on pseudo-labels generated by a Hidden Markov Model (HMM), b) DNN fine-tuning integrating the annotations of an expert Ob-Gyn clinician. The trained DNN exhibits promising results: Balanced Accuracy of 88.37%, Macro F1-Score of 87.87% and Matthews Correlation Coefficient (MCC) of 75.80% on a distinct hold-out test set, encompassing 45 FHR traces annotated by an expert Ob-Gyn clinician.

Automated Driving (AD) technologies are rapidly transforming road transportation, emphasizing the critical role of Human-Machine Interaction (HMI). In this regard, the paper examines the interaction between Level 4 Autonomous Vehicles (L4 AVs) and human drivers in take-over scenarios within Italian traffic environments. Employing the Dynamic Driving Simulator at Politecnico di Milano, the study presents two simulation environments: an urban roundabout and a Ligurian highway. The research aims to measure the driver response during take-over requests. Questionnaires are used to psychologically analyse the participants. Physiological signals, including ECG, EEG, and EDA, are acquired throughout the entire simulation.

Emotions play a critical role in shaping our daily experiences by influencing decision-making, perception, learning, thinking, and behavior. Electroencephalographic (EEG) recordings are commonly utilized to measure brain activity, offering valuable insights into emotional states and allowing researchers to gain deeper understanding of the mechanisms underlying affective processes. The goals of this study are to establish a protocol using validated stimuli from literature to evaluate the effectiveness of EEG signals on an individual level for classifying arousal and valence of elicited stimuli and to identify the most effective feature space, brain regions, and signal frequencies for this purpose. To this extent, specific experimental procedures were employed to elicit four emotions, and standardized preprocessing, feature extraction, selection, and classification processes across subjects were applied to evaluate affective decoding. Classification of arousal and valence levels as binary tasks achieved highest accuracy when utilizing frequency domain features, with 80.1% for arousal and 74.8% for valence. Frequency domain features proved particularly effective in distinguishing varying levels of arousal and valence. The frontal and central cortices emerged as crucial in cognitive processing related to emotion, with higher EEG frequency bands closely associated with emotion processing. Nonetheless, the delta band also showed significance in classification and warrants further exploration, particularly regarding arousal. The study refines EEG-based arousal and valence classification, highlighting optimal feature space, brain regions, and frequency bands, thereby advancing understanding of affective processes.

The current scenario of the Architecture, Engineering, Construction and Operation (AECO) industry outlines an accelerating approach to digitalisation which is defined PropTech (abbreviation for Property Technologies). Among all the digital innovation brought to AECO, PropTech companies are introducing digital technologies in the Operation and Maintenance (O&M) phase of the building lifecycle to improve operational efficiency, performance of an asset, and effectiveness of provided services and supplies. Analysing the literature scenario of O&M phase, Building Information Modelling (BIM) and Digital Twin (DT) have resulted in a methodological innovation for the entire industry. Thanks to Internet of Things (IoT), the advent of DT makes its way into the building sector allowing among all to monitor the as-is conditions, detect anomalies before they occur, make diagnosis, and give an added value with respect to the BIM. Although DT is in its early stages, in the real estate market, some PropTech companies have embraced the challenge and applied the digital technology for building management. Therefore, the paper aims to analyse the numerous advantages of DT in the management of buildings. For this objective, the authors referred to two case studies, taken from the Italian PropTech Network ecosystem, that implement DT in management of O&M phase. On one hand the authors present the benefits of facility management digitalisation introduced by DT; on the other, they underline the issues faced by the two companies and the future implementations of DT in the O&M field.

EDITORIAL article Front. Bioeng. Biotechnol., 01 August 2023Sec. Biosensors and Biomolecular Electronics Volume 11 - 2023 | https://doi.org/10.3389/fbioe.2023.1239031

Cardiotocography (CTG) is the most common technique for electronic fetal monitoring and consists of the simultaneous recording of fetal heart rate (FHR) and uterine contractions. In analogy with the adult case, spectral analysis of the FHR signal can be used to assess the functionality of the autonomic nervous system. To do so, several methods can be employed, each of which has its strengths and limitations. This paper aims at performing a methodological investigation on FHR spectral analysis adopting 4 different spectrum estimators and a novel PRSA-based spectral method. The performances have been evaluated in terms of the ability of the various methods to detect changes in the FHR in two common pregnancy complications: intrauterine growth restriction (IUGR) and gestational diabetes. A balanced dataset containing 2178 recordings distributed between the 32nd and 38th week of gestation was used. The results show that the spectral method derived from the PRSA better differentiates high-risk pregnancies vs. controls compared to the others. Specifically, it more robustly detects an increase in power percentage within the movement frequency band and a decrease in high frequency between pregnancies at high risk in comparison to those at low risk.

Background diabetes is a very common pregnancy complication. This study aims to compare measurements taken during antenatal monitoring in the third trimester and delivery outcome data in pregnancies complicated by maternal diabetes with respect to healthy controls. Methods a prospective observational study included a total of 152 pregnant women. 25 had a diagnosis of pregestational diabetes mellitus (PGDM) and 61 were diagnosed with gestational diabetes mellitus (GDM). The remaining 66 were controls. Data collected antepartum for all pregnancies include ultrasound fetal biometry and amniotic fluid evaluation, Umbilical and Middle Cerebral Artery Pulsatility Index (UA-PI, MCA-PI) and computerized CTG (cCTG) monitoring. The mode of delivery and neonatal data, including umbilical cord gas values, were also obtained. The values collected were compared between groups. Results GDM and PGDM groups showed significantly lower values of UA-PI compared to Controls (p < 0.001) and GDMs assumed slightly lower values of MCA-PI compared to Controls (p = 0.028). Some cCTG parameters showed small but significant differences among groups. The analyzed groups presented significant differences in the umbilical artery gas analysis values at birth, which were affected by the mode of delivery. The pH was lower in PGDMs compared to both GDMs (p = 0.0279) and Controls (p < 0.0001), and spontaneous deliveries were associated with lower pH values (p = 0.008). pO2 significantly decreased from Controls to GDMs and PGDMs, respectively (Controls vs GDMs p = 0.0057, Controls vs PGDMs p < 0.001, GDMs vs PGDMs p = 0.0002) while pCO2 followed an opposite trend (Controls vs GDMs p < 0.001, Controls vs PGDMs p < 0.001, GDMs vs PGDMs p = 0.0014). Lactates were higher in PGDMs compared to both other groups (Controls vs PGDMs: p = 0.0128, GDMs vs PGDMs: p = 0.0161) and were higher in vaginal deliveries compared to cesarean sections (p = 0.017). Conclusions The results suggest that to date there are no antenatal monitoring methods that can accurately define the true well-being of the fetus in pregnancies complicated by diabetes. In fact, neonatal outcome data show greater differences between groups than those identified antepartum by the analyzed methods. We hypothesize that only a multi-parametric/multimodal approach can help in antepartum management.

BACKGROUND AND OBJECTIVES:Computerized Cardiotocography (cCTG) allows to analyze the Fetal Heart Rate (FHR) objectively and thoroughly, providing valuable insights on fetal condition. A challenging but crucial task in this context is the automatic identification of fetal activity and quiet periods within the tracings. Different neural mechanisms are involved in the regulation of the fetal heart, depending on the behavioral states. Thereby, their correct identification has the potential to increase the interpretability and diagnostic capabilities of FHR quantitative analysis. Moreover, the most common pathologies in pregnancy have been associated with variations in the alternation between quiet and activity states.METHODS:We address the problem of fetal states clustering by means of an unsupervised approach, resorting to the use of a multivariate Hidden Markov Models (HMM) with discrete emissions. A fixed length sliding window is shifted on the CTG traces and a small set of features is extracted at each slide. After an encoding procedure, these features become the emissions of a multivariate HMM in which quiet and activity are the hidden states. After an unsupervised training procedure, the model is used to automatically segment signals.RESULTS:The achieved results indicate that our developed model exhibits a high degree of reliability in identifying quiet and activity states within FHR signals. A set of 35 CTG signals belonging to different pregnancies were independently annotated by an expert gynecologist and segmented using the proposed HMM. To avoid any bias, the physician was blinded to the results provided by the algorithm. The overall agreement between the HMM's predictions and the clinician's interpretations was 90%.CONCLUSIONS:The proposed method reliably identified fetal behavioral states, the alternance of which is an important factor in the fetal development. One key strength of our approach lies in the ease of interpreting the obtained results. By utilizing a small set of parameters that are already used in cCTG and possess clear intrinsic meanings, our method provides a high level of explainability. Another significant advantage of our approach is its fully unsupervised learning process. The states identified by our model using the Baum-Welch algorithm are associated with the "Active" and "Quiet" states only after the clustering process, removing the reliance on expert annotations. By autonomously identifying the clusters based solely on the intrinsic characteristics of the signal, our method achieves a more objective evaluation that overcomes the limitations of subjective interpretations. Indeed, we believe it could be integrated in cCTG systems to obtain a more complete signal analysis.

In pregnancy, diabetes is known to increase the risk of adverse maternal and neonatal outcomes.It would be beneficial to find techniques that allow early investigation of the physio-pathological mechanisms involved to provide clinicians with tools for prevention and therapies.For that, cardiotocography (CTG) is a promising tool.However, the evidence is still scarce and the impact on clinical practice little.In this study, we aim at characterizing the changes induced by gestational diabetes (GDM) on the fetal heart rate series.To do so, we performed a retrospective cohort study on a CTG dataset containing more than 20000 recordings of which 852 belong to 301 GDM-diagnosed patients.We divided the recordings by gestational age (G.A.) into 4 groups (weeks: 31-35, 36, 37, 38 to delivery) and for each we identified a control population of equal size matched by comorbidities.We analyzed a comprehensive set of parameters from the time domain, frequency domain and non-linear analysis and assessed variations in median values on each feature.For all G.A. below the 38th week, we found a significant increase in the power in the movement frequency band (p<0.01) and an increase in the absolute value of Deceleration Reserve (p<0.01) in GDM vs control.Other significant values were also identified and are discussed in more detail in the paper.

Abstract Objectives To investigate the use of computerized cardiotocography (cCTG) parameters in Intrauterine Growth Restriction (IUGR) pregnancies for the prediction of 1) complication with preeclampsia; 2) placental histological abnormalities, and 3) neonatal outcomes.  Study design A single-center observational retrospective case-control study was performed by reviewing medical records, cCTG databases and pathological reports of women with singleton pregnancy and IUGR uncomplicated (controls) and complicated by preeclampsia (cases). Primary endpoint was the association between cCTG parameters and preeclampsia in IUGR. Secondary endpoints were the association between cCTG parameters and 1) placental abnormalities, and 2) neonatal outcomes. The one-way ANOVA test was used to compare cCTG parameters in cases and controls. t-test was applied to compare neonatal outcomes and placental abnormalities in both groups. The Spearman Test value Correlation coefficients between the cCTG parameters and neonatal outcome in the two groups. A p value < .05 was considered significant for all analyses. Results Among all cCTG parameters, a significant association with preeclampsia in IUGR was found for Fetal Heart Rate (FHR, p = .008), Delta (p = .018), Short Term Variability (STV, p = .021), Long Term Variability (LTV, p = .028), Acceleration Phase Rectified Slope (APRS, p = .018) and Deceleration Phase Rectified Slope (DPRS, p = .038). Of all placental histologic abnormalities, only vascular alterations at least moderate were significantly associated with increased FHR (p = .02). About neonatal outcomes, all cCTG parameters were significantly associated with birth weight, Apgar index at 1 and 5 min, pH and pCO2. FHR, LTI, Delta, Approximate Entropy (ApEn) and LF were significantly associated with pO2; LTI, Interval Index (II) and ApEn with base excess. Among controls, Delta, ApEn, Low Frequency (LF) and High Frequency (HF) were significantly associated with pCO2, while among cases, STV and Delta were significantly associated with pH; STV, LTI, Delta, ApEn, LF and HF with pCO2; STV, LTI, Delta and ApEn with pO2; HF with base excess; FHR and LF with lactates. Conclusions cCTG parameters may be useful to detect complication with preeclampsia in IUGR pregnancies. Regarding placental status, cCTG parameters may detect overall circulation alterations, but not specific histological abnormalities. Lastly, all cCTG parameters may predict neonatal outcomes, helping to tailor the patients’ management.

The Cardiotocography (CTG) is a widely diffused monitoring practice, used in Ob-Gyn Clinic to assess the fetal well-being through the analysis of the Fetal Heart Rate (FHR) and the Uterine contraction signals. Due to the complex dynamics regulating the Fetal Heart Rate, a reliable visual interpretation of the signal is almost impossible and results in significant subjective inter and intra-observer variability. Also, the introduction of few parameters obtained from computer analysis did not solve the problem of a robust antenatal diagnosis. Hence, during the last decade, computer aided diagnosis systems, based on artificial intelligence (AI) machine learning techniques have been developed to assist medical decisions. The present work proposes a hybrid approach based on a neural architecture that receives heterogeneous data in input (a set of quantitative parameters and images) for classifying healthy and pathological fetuses. The quantitative regressors, which are known to represent different aspects of the correct development of the fetus, and thus are related to the fetal healthy status, are combined with features implicitly extracted from various representations of the FHR signal (images), in order to improve the classification performance. This is achieved by setting a neural model with two connected branches, consisting respectively of a Multi-Layer Perceptron (MLP) and a Convolutional Neural Network (CNN). The neural architecture was trained on a huge and balanced set of clinical data (14.000 CTG tracings, 7000 healthy and 7000 pathological) recorded during ambulatory non stress tests at the University Hospital Federico II, Napoli, Italy. After hyperparameters tuning and training, the neural network proposed has reached an overall accuracy of 80.1%, which is a promising result, as it has been obtained on a huge dataset.

The purpose of this study is to develop and understand whether Machine Learning models can classify Cardiotocographic (CTG) recordings of healthy fetuses or Intra Uterine Growth Restricted (IUGR) fetuses, highlighting how a large amount of data can have unexpected effects. We started from other findings in the literature to see what Machine Learning model remained consistent even with a large amount of data. The CTG records used in this study were collected at the Department of Obstetrics of the Federico II University Hospital in Naples, Italy, from 2013 to 2021. From this dataset, we chose 1548 IUGR fetuses and 1548 healthy fetuses to train our models. Each recording contained several parameters, ranging from features calculated on the entire CTG tracing, features calculated every 3 and 1 minute of recording and features related to the pregnant woman, such as age and week of gestation. We trained our machine-learning models on this dataset, checking the results obtained before and after adjusting the hyperparameters, noting that among the best models was Random Forest, which has already been present in other studies, and that the Multilayer Perceptron and the AdaBoost classifier were overall the best performing. This work can surely form a basis for future works in the fetal heart rate classification thus leading to real clinical applications.

In this work we present the creation of a large, structured database of CardioTocoGraphic (CTG) recordings, starting from a raw dataset containing tracings collected between 2013 and 2021 by the medical team of the University Hospital Federico II of Naples. The aim of the work is to provide a big, structured database of real clinical cardiotocographic data, useful for subsequent processing and analysis through state-of-the-art methods, in particular Deep Learning Methods. This organized dataset could lead to an increase of the diagnostic accuracy of CTG analysis in the discrimination of healthy and unhealthy fetuses.

Abstract Background The clinical diagnosis of late Fetal Growth Restriction (FGR) involves the integration of Doppler ultrasound data and Fetal Heart Rate (FHR) monitoring through computer assisted computerized cardiotocography (cCTG). The aim of the study was to evaluate the diagnostic power of combined Doppler and cCTG parameters by contrasting late FGR –and healthy controls. Methods The study was conducted from January 2018 to May 2020. Only pregnant women who had the last Doppler measurement obtained within 1 week before delivery and cCTG performed within 24 h before delivery were included in the study. Two hundred forty-nine pregnant women fulfilling the inclusion criteria were enrolled in the study; 95 were confirmed as late FGR and 154 were included in the control group. Results Among the extracted cCTG parameters, Delta Index, Short Term Variability (STV), Long Term Variability (LTV), Acceleration and Deceleration Phase Rectified Slope (APRS, DPRS) values were lower in the late FGR participants compared to the control group. In the FGR cohort, Delta, STV, APRS, and DPRS were found different when stratifying by MCA_PI (MCA_PI <5th centile or > 5th centile). STV and DPRS were the only parameters to be found different when stratifying by (UA_PI >95th centile or UA_PI <95th centile). Additionally, we measured the predictive power of cCTG parameters toward the identification of associated Doppler measures using figures of merit extracted from ROC curves. The AUC of ROC curves were accurate for STV (0,70), Delta (0,68), APRS (0,65) and DPRS (0,71) when UA_PI values were > 95th centile while, the accuracy attributable to the prediction of MCA_PI was 0.76, 0.77, 0.73, and 0.76 for STV, Delta, APRS, and DPRS, respectively. An association of UA_PI>95th centile and MCA_PI<5th centile with higher risk for NICU admission, was observed, while CPR < 5th centile resulted not associated with any perinatal outcome. Values of STV, Delta, APRS, DPRS were significantly lower for FGR neonates admitted to NICU, compared with the uncomplicated FGR cohort. Conclusions The results of this study show the contribution of advanced cCTG parameters and fetal Doppler to the identification of late FGR and the association of those parameters with the risk for NICU admission. Trial registration Retrospectively registered.

The well known approach for the quantification of the sympatho-vagal balance controlling heart rate, fails when the applications hypothesis are not fulfilled. Different models are introduced for a better characterization of the cardiorespiratory interactions in different conditions.