Search

Emotion recognition is a challenging task with many potential applications in psychology, psychiatry, and human-computer interaction (HCI). The use of time-delay information in the controlled time-delay stability (cTDS) algorithm can help to capture the temporal dynamics of EEG signals, including sub-band information and bi-directional coupling that can aid in emotion recognition and identification of specific connectivity patterns between brain rhythms. Incorporating EEG frequency bands can be used to design better emotion recognition systems. This paper evaluates the cTDS algorithm for binary classification tasks of arousal and valence using EEG sub-band signals. This method achieved a high accuracy of 91.1% for arousal and 91.7% for valence based on one electrode recording site at Fp1. The cTDS algorithm is a promising approach to analyzing brain network interactions. It can be particularly applicable to arousal and valence classification tasks, especially within a complex, multimodal feature space associated with understanding psychiatric disorders and HCI applications.

This paper explores automated face and facial landmark detection of neonates, which is an important first step in many video-based neonatal health applications, such as vital sign estimation, pain assessment, sleep-wake classification, and jaundice detection. Utilising three publicly available datasets of neonates in the clinical environment, 366 images (258 subjects) and 89 (66 subjects) were annotated for training and testing, respectively. Transfer learning was applied to two YOLO-based models, with input training images augmented with random horizontal flipping, photo-metric colour distortion, translation and scaling during each training epoch. Additionally, the re-orientation of input images and fusion of trained deep learning models was explored. Our proposed model based on YOLOv7Face outperformed existing methods with a mean average precision of 84.8% for face detection, and a normalised mean error of 0.072 for facial landmark detection. Overall, this will assist in the development of fully automated neonatal health assessment algorithms.Clinical relevance- Accurate face and facial landmark detection provides an automated and non-contact option to assist in video-based neonatal health applications.

Neonatal respiratory distress is a common condition that if left untreated, can lead to short- and long-term complications. This paper investigates the usage of digital stethoscope recorded chest sounds taken within 1 min post-delivery, to enable early detection and prediction of neonatal respiratory distress. Fifty-one term newborns were included in this study, 9 of whom developed respiratory distress. For each newborn, 1 min anterior and posterior recordings were taken. These recordings were pre-processed to remove noisy segments and obtain high-quality heart and lung sounds. The random undersampling boosting (RUSBoost) classifier was then trained on a variety of features, such as power and vital sign features extracted from the heart and lung sounds. The RUSBoost algorithm produced specificity, sensitivity, and accuracy results of 85.0%, 66.7% and 81.8%, respectively. Clinical relevance--- This paper investigates the feasibility of digital stethoscope recorded chest sounds for early detection of respiratory distress in term newborn babies, to enable timely treatment and management.

Obtaining high quality heart and lung sounds enables clinicians to accurately assess a newborns cardio-respiratory health and provide timely care. However, noisy chest sound recordings are common, hindering timely and accurate assessment. A new Non-negative Matrix Co-Factorisation based approach is proposed to separate noisy chest sound recordings into heart, lung and noise components to address this problem. This method is achieved through training with 20 high quality heart and lung sounds, in parallel with separating the sounds of the noisy recording. The method was tested on 68 10-second noisy recordings containing both heart and lung sounds and compared to the current state of the art Non-negative Matrix Factorisation methods. Results show significant improvements in heart and lung sound quality scores respectively, and improved accuracy of 3.6bpm and 1.2bpm in heart and breathing rate estimation respectively, when compared to existing methods.

Fetal myocardial performance index are used to assess aspects of systolic and diastolic function in developing fetal heart. The aim of this study was to determine normal values of fetal left ventricular (LV) Tei Index (TI) for measuring systolic performance and K-Index (KI) for diastolic performance in early (<;32 weeks), Mid (32-35 weeks) and late trimester (35-41 weeks) normal fetuses and fetuses with congenital heart diseases (CHD). Simultaneous recordings of Doppler Ultrasound Signal (DUS) of the LV outflow tracts and fetal electrocardiogram signals were obtained in 57 normal and 14 fetuses with CHD. CHDs were categorized in two types (conduction pathway abnormalities [8 cases] and structural anomalies [6 cases]). The LV isovolumic contraction time (ICT), isovolumic relaxation time (IRT), ventricular ejection time (VET) and ventricular filling time (VFT) were measured from amplitudes and peaks of high frequency component of DUS signals by an automated model. The TI and the new index were calculated by using the formula (ICT + IRT)/VET and (ICT + IRT)/VFT respectively. The TI did not show any significant change from CHD (conduction) to CHD (structural) fetuses (0.72± 0.09 vs 0.76±0.02; NS). On the other hand, KI showed significant decline in values (0.82±0.24 vs 0.49±0.34; p<;0.05) during normal development period and CHD with structural anomalies particularly. The correlation of KI with gestational ages was found to be negative and significant (r=- 0.29; p=0.025). Both TI and KI can be easily obtained in the fetus from DUS recording without the need for precise anatomic imaging. In conclusion, KI may be a useful tool to explore the healthy development of fetal myocardial function and identify fetuses with CHD (structural). More research is needed to check if KI could recognize the compromised fetuses from normal ones in future.

Doppler-derived myocardial performance index are used to assess aspects of systolic and diastolic function in adult as well as fetal heart. The Tei Index (TI) is a useful and non-invasive tool. The aim of this study was to determine normal values of fetal left ventricular (LV) TI in early (16-32 weeks) and late trimester (35-41 weeks) fetuses and to develop a new myocardial performance index (K-index; KI) which focuses on fetal diastolic interval when oxygenated blood from the placenta filled in the LV. Simultaneous recordings of Doppler Ultrasound Signal (DUS) of the LV outflow tracts and fetal electrocardiogram signals were obtained in 55 early- and late-trimester fetuses. The LV isovolumic contraction time (ICT), isovolumic relaxation time (IRT), ventricular ejection time (VET) and ventricular filling time (VFT) were measured from amplitudes and peaks of high frequency component of DUS signals by an automated model. The TI and the new index were calculated by using the formula (ICT + IRT)/VET and (ICT + IRT)/VFT respectively. The TI did not show any significant change from early-to late trimester fetuses (0.74± 0.057 vs 0.75±0.07; NS). On the other hand, KI showed significant decline in values (1.00±0.24 vs 0.87±0.23; p<;0.05) during this development period. The correlation of KI with gestational ages was found to be negative and significant (r=-0.35; p=0.007). No significant correlation of gestational ages with TI was found. Both TI and KI can be easily obtained in the fetus from DUS recording without the need for precise anatomic imaging. In conclusion, KI may be a useful tool to explore the healthy development of fetal myocardial function. More research is needed to check if KI could recognize the compromised fetuses from normal ones in future.

Although evidence of the short term relationship between maternal and fetal heart rates has been found in previous model-based studies, knowledge about the mechanism and patterns of the coupling during gestation is still limited. In this study, a model-free method based on Transfer Entropy (TE) was applied to quantify the maternal-fetal heart rate couplings in both directions. Furthermore, analysis of the lag at which TE was maximum and its changes throughout gestation, provided more information about the mechanism of coupling and its latency. Experimental results based on fetal electrocardiograms (fECGs) and maternal ECG showed the evidence of coupling for 62 out of 65 healthy mothers and fetuses in each direction, by statistically validating against the surrogate pairs. The fetuses were divided into three gestational age groups: early (16-25 weeks), mid (26-31 weeks) and late (32-41 weeks) gestation. The maximum TE from maternal to fetal heart rate significantly increased from early to mid gestation, while the coupling delay on both directions decreased significantly from mid to late gestation. These changes occur concomitant with the maturation of the fetal sensory and autonomic nervous systems with advancing gestational age. In conclusion, the application of TE with delays revealed detailed information about the changes in fetal-maternal heart rate coupling strength and latency throughout gestation, which could provide novel clinical markers of fetal development and well-being.

Heart Rate Variability (HRV) is extensively used to investigate general Autonomic Nervous System (ANS) function and is affected by many factors including age, gender, pathology such as diabetes and genetic polymorphisms. One of these genetic polymorphisms is the Angiotensin Converting Enzyme (ACE) polymorphism corresponding to insertion (I) or deletion (D) of a 287-base pair sequence of DNA in intron 16 of the ACE gene (rs4340). Some studies have addressed the relationship between HRV and D/D, I/D and I/I ACE polymorphism while others combined I/D and I/I ACE groups. In this study HRV is determined for diabetic and control individuals with different ACE polymorphism considering either separate or combined I/D and I/I genotypes. Linear time domain parameters, entropy, low frequency and total power of HRV were found to be significantly different between diabetic and control individuals with combined I/I and I/D ACE polymorphism, while only entropy was different for diabetic and control subjects with D/D ACE genotype. Separate analysis of I/I and I/D genotypes was preferred for a thorough investigation of HRV and ACE polymorphism, as the combined analysis masked some differences in HRV parameters such as Poincaré plot between ACE polymorphisms and diabetes status. Furthermore, a separate analysis demonstrated that most of the significant differences for HRV were between the diabetic group with I/I genotype and I/D and D/D groups.

Fetal cardiac assessment techniques are aimed to identify fetuses at risk of intrauterine compromise or death. Evaluation of the electromechanical coupling as a fundamental part of the fetal heart physiology, provides valuable information about the fetal wellbeing during pregnancy. It is based on the opening and closing time of the cardiac valves and the onset of the QRS complex of the fetal electrocardiogram (fECG). The focus of this paper is on the automated identification of the fetal cardiac valve opening and closing from Doppler Ultrasound signal and fECG as a reference. To this aim a novel combination of Emprical Mode Decomposition (EMD) and multi-dimensional Hidden Markov Models (MD-HMM) was employed which provided beat-to-beat estimation of cardiac valve event timings with improved precision (82.9%) compared to the one dimensional HMM (77.4%) and hybrid HMM-Support Vector Machine (SVM) (79.8%) approaches.

The development of the fetal cardiovascular system plays a crucial role in fetal health. The evolution of the relationship between fetal and maternal cardiac systems during fetal maturation is a characterizing feature for fetal cardiac development. This paper aims to evaluate this relationship by investigating the beat-to-beat synchronization between fetal and maternal heart rates and its variation at different stages of pregnancy. Synchronization epochs and phase locking patterns are analyzed at certain synchronization ratios (SRs) for three gestational age groups (16-26 weeks, 27-33 weeks, 34-40 weeks). Results show that the normalized synchronization epoch is significantly different for three age groups with the p-value of 6.72*10(-6) and 2.89*10(-4) at SR of 1:2 and 4:5 respectively. The variance of phase locking also shows significant difference for three groups with the p-value less than 10(-7) at four SRs. Results also suggest that synchronization may be the force behind the increase in the maternal heart rate to maintain the fetal development and provide supplies for the fetus. Overall, the findings propose new clinical markers for evaluating the antenatal development.

In this paper a new noninvasive method is proposed for automated estimation of opening and closure timings of fetal cardiac valves. These timings are obtained from Doppler Ultrasound (DUS) signal and fetal electrocardiogram (fECG) as a reference. Empirical Mode Decomposition (EMD) is first applied to the DUS signal to decompose it into different components called Intrinsic Mode Functions (IMFs). The envelope of the first IMF is then taken and its peaks are identified. The opening and closure of the valves are then automatically assigned to the IMF peaks by using Hidden Markov Model (HMM). It is shown that this new method can continuously evaluate fetal cardiac valves' (aortic and mitral) motion timings for 82.5~99.7% of cardiac cycles. The estimated timings are verified using the Pulsed Doppler images. These findings can be used as sensitive markers for evaluating the fetal cardiac performance.