Circadian assessment of heart failure using explainable deep learning and novel multi-parameter polar images

BACKGROUND AND OBJECTIVE

Heart failure (HF) is a multi-faceted and life-threatening syndrome that affects more than 64.3 million people worldwide. Current gold-standard screening technique, echocardiography, neglects cardiovascular information regulated by the circadian rhythm and does not incorporate knowledge from patient profiles. In this study, we propose a novel multi-parameter approach to assess heart failure using heart rate variability (HRV) and patient clinical information.

METHODS

In this approach, features from 24-hour HRV and clinical information were combined as a single polar image and fed to a 2D deep learning model to infer the HF condition. The edges of the polar image correspond to the timely variation of different features, each of which carries information on the function of the heart, and internal illustrates color-coded patient clinical information.

RESULTS

Under a leave-one-subject-out cross-validation scheme and using 7,575 polar images from a multi-center cohort (American and Greek) of 303 coronary artery disease patients (median age: 58 years [50-65], median body mass index (BMI): 27.28 kg/m2 [24.91-29.41]), the model yielded mean values for the area under the receiver operating characteristics curve (AUC), sensitivity, specificity, normalized Matthews correlation coefficient (NMCC), and accuracy of 0.883, 90.68%, 95.19%, 0.93, and 92.62%, respectively. Moreover, interpretation of the model showed proper attention to key hourly intervals and clinical information for each HF stage.

CONCLUSIONS

The proposed approach could be a powerful early HF screening tool and a supplemental circadian enhancement to echocardiography which sets the basis for next-generation personalized healthcare.

Identification of Congenital Valvular Murmurs in Young Patients Using Deep Learning-Based Attention Transformers and Phonocardiograms

One in every four newborns suffers from congenital heart disease (CHD) that causes defects in the heart structure. The current gold-standard assessment technique, echocardiography, causes delays in the diagnosis owing to the need for experts who vary markedly in their ability to detect and interpret pathological patterns. Moreover, echo is still causing cost difficulties for low- and middle-income countries. Here, we developed a deep learning-based attention transformer model to automate the detection of heart murmurs caused by CHD at an early stage of life using cost-effective and widely available phonocardiography (PCG). PCG recordings were obtained from 942 young patients at four major auscultation locations, including the aortic valve (AV), mitral valve (MV), pulmonary valve (PV), and tricuspid valve (TV), and they were annotated by experts as absent, present, or unknown murmurs. A transformation to wavelet features was performed to reduce the dimensionality before the deep learning stage for inferring the medical condition. The performance was validated through 10-fold cross-validation and yielded an average accuracy and sensitivity of 90.23 % and 72.41 %, respectively. The accuracy of discriminating between murmurs' absence and presence reached 76.10 % when evaluated on unseen data. The model had accuracies of 70 %, 88 %, and 86 % in predicting murmur presence in infants, children, and adolescents, respectively. The interpretation of the model revealed proper discrimination between the learned attributes, and AV channel was found important (score 0.75) for the murmur absence predictions while MV and TV were more important for murmur presence predictions. The findings potentiate deep learning as a powerful front-line tool for inferring CHD status in PCG recordings leveraging early detection of heart anomalies in young people. It is suggested as a tool that can be used independently from high-cost machinery or expert assessment.

Investigating automated regression models for estimating left ventricular ejection fraction levels in heart failure patients using circadian ECG features

Heart Failure (HF) significantly impacts approximately 26 million people worldwide, causing disruptions in the normal functioning of their hearts. The estimation of left ventricular ejection fraction (LVEF) plays a crucial role in the diagnosis, risk stratification, treatment selection, and monitoring of heart failure. However, achieving a definitive assessment is challenging, necessitating the use of echocardiography. Electrocardiogram (ECG) is a relatively simple, quick to obtain, provides continuous monitoring of patient's cardiac rhythm, and cost-effective procedure compared to echocardiography. In this study, we compare several regression models (support vector machine (SVM), extreme gradient boosting (XGBOOST), gaussian process regression (GPR) and decision tree) for the estimation of LVEF for three groups of HF patients at hourly intervals using 24-hour ECG recordings. Data from 303 HF patients with preserved, mid-range, or reduced LVEF were obtained from a multicentre cohort (American and Greek). ECG extracted features were used to train the different regression models in one-hour intervals. To enhance the best possible LVEF level estimations, hyperparameters tuning in nested loop approach was implemented (the outer loop divides the data into training and testing sets, while the inner loop further divides the training set into smaller sets for cross-validation). LVEF levels were best estimated using rational quadratic GPR and fine decision tree regression models with an average root mean square error (RMSE) of 3.83% and 3.42%, and correlation coefficients of 0.92 (p<0.01) and 0.91 (p<0.01), respectively. Furthermore, according to the experimental findings, the time periods of midnight-1 am, 8-9 am, and 10-11 pm demonstrated to be the lowest RMSE values between the actual and predicted LVEF levels. The findings could potentially lead to the development of an automated screening system for patients with coronary artery disease (CAD) by using the best measurement timings during their circadian cycles.

Non-invasive technologies for heart failure, systolic and diastolic dysfunction modeling: a scoping review

The growing global prevalence of heart failure (HF) necessitates innovative methods for early diagnosis and classification of myocardial dysfunction. In recent decades, non-invasive sensor-based technologies have significantly advanced cardiac care. These technologies ease research, aid in early detection, confirm hemodynamic parameters, and support clinical decision-making for assessing myocardial performance. This discussion explores validated enhancements, challenges, and future trends in heart failure and dysfunction modeling, all grounded in the use of non-invasive sensing technologies. This synthesis of methodologies addresses real-world complexities and predicts transformative shifts in cardiac assessment. A comprehensive search was performed across five databases, including PubMed, Web of Science, Scopus, IEEE Xplore, and Google Scholar, to find articles published between 2009 and March 2023. The aim was to identify research projects displaying excellence in quality assessment of their proposed methodologies, achieved through a comparative criteria-based rating approach. The intention was to pinpoint distinctive features that differentiate these projects from others with comparable objectives. The techniques identified for the diagnosis, classification, and characterization of heart failure, systolic and diastolic dysfunction encompass two primary categories. The first involves indirect interaction with the patient, such as ballistocardiogram (BCG), impedance cardiography (ICG), photoplethysmography (PPG), and electrocardiogram (ECG). These methods translate or convey the effects of myocardial activity. The second category comprises non-contact sensing setups like cardiac simulators based on imaging tools, where the manifestations of myocardial performance propagate through a medium. Contemporary non-invasive sensor-based methodologies are primarily tailored for home, remote, and continuous monitoring of myocardial performance. These techniques leverage machine learning approaches, proving encouraging outcomes. Evaluation of algorithms is centered on how clinical endpoints are selected, showing promising progress in assessing these approaches' efficacy.

Classification of children with ADHD through task-related EEG recordings via Swarm-Decomposition-based Phase Locking Value. Annu Int Conf IEEE Eng Med Biol Soc 2023;2023:1-5. [PMID: 38082916 DOI: 10.1109/embc40787.2023.10340329] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

Attention Deficit/Hyperactivity Disorder (ADHD) is a common neurodevelopmental disorder mainly affecting children. ADHD children brain activity is reported to present alterations from neurotypically developed children, yet establishment of an EEG biomarker, which is of high importance in clinical practice and research, has not been achieved. In this work, task-related EEG recordings from 61 ADHD and 60 age-matched non-ADHD children are analyzed to examine the underlying Cross-Frequency Coupling phenomena. The proposed framework introduces personalized brain rhythm extraction in the form of oscillatory modes via Swarm Decomposition, allowing for the transition from sensor-level connectivity to source-level connectivity. Oscillatory modes are then subjected to a phase locking value-based feature extraction and the efficiency of the extracted features in separating ADHD from non-ADHD individuals is evaluated by means of a nested 5-fold cross validation scheme. The experimental results of the proposed framework (Area Under the Receiver Operating Characteristics Curve-AUROC: 0.9166) when benchmarked against the commonly used filter-based brain rhythm extraction (AUROC: 0.8361) underscore its efficiency and demonstrate its overall superiority over other state-of-the-art functional connectivity approaches in this classification task for this dataset.Clinical relevance-This framework provides novel insights about brain regions of interest that are involved in ADHD task-related function and holds promise in providing objective ADHD biomarkers by extending classic sensor-level connectivity to source-level.

Emotional Climate Recognition in Conversations using Peers' Speech-based Bispectral Features and Affect Dynamics. Annu Int Conf IEEE Eng Med Biol Soc 2023;2023:1-5. [PMID: 38083331 DOI: 10.1109/embc40787.2023.10340028] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

Emotion recognition in conversations using artificial intelligence (AI) has recently gained a lot of attention, as it can provide additional emotion cues that can be correlated with human social behavior. An extension towards an AI-based emotional climate (EC) recognition, i.e., the recognition of the joint emotional atmosphere dynamically created and perceived by the peers throughout a conversation, is proposed here. In our approach, namely MLBispeC (Machine Learning Based Bispectral Classification), the peers' speech signals during their conversation are subjected to time-windowed bispectral analysis, allowing for feature extraction related to dynamic harmonics nonlinear interactions. In addition, peers' affect dynamics, derived from their same time-windowed emotion labeling, are combined to form an extended feature vector, inputted into two well-known machine learning classifiers (Support Vector Machine, K-Nearest Neighbor). MLBispeC was evaluated on the Interactive Emotional Dyadic Motion Capture (IEMOCAP) open access dataset, which contains 2D emotions, i.e., Arousal (A) and valence (V) that are divided into (low/high) classes. The experimental results have shown that MLBispeC outperforms previous state-of-the-art techniques, achieving an accuracy of 0.826A/0.754V, sensitivity of 0.864A/0.774V, and area under the curve (AUC) of 0.821A/0.799V. This demonstrates the effectiveness of MLBispeC to objectively recognize peers' EC during their conversation, allowing for insights into their emotional and social interactions.Clinical relevance-Unobtrusive, objective and dynamic recognition of the EC built during peers' conversation can scaffold effective assessment of patients with physiological, psychological, and mental diseases, at various age ranges (children, adults, and older adults).

Heart Failure Assessment Using Multiparameter Polar Representations and Deep Learning. Annu Int Conf IEEE Eng Med Biol Soc 2023;2023:1-4. [PMID: 38083567 DOI: 10.1109/embc40787.2023.10341132] [Citation(s) in RCA: 0] [Impact Index Per Article: 0]

Heart failure refers to the inability of the heart to pump enough amount of blood to the body. Nearly 7 million people die every year because of its complications. Current gold-standard screening techniques through echocardiography do not incorporate information about the circadian rhythm of the heart and clinical information of patients. In this vein, we propose a novel approach to integrate 24-hour heart rate variability (HRV) features and patient profile information in a single multi-parameter and color-coded polar representation. The proposed approach was validated by training a deep learning model from 7,575 generated images to predict heart failure groups, i.e., preserved, mid-range, and reduced left ventricular ejection fraction. The developed model had overall accuracy, sensitivity, and specificity of 93%, 88%, and 95%, respectively. Moreover, it had a high area under the receiver operating characteristics curve (AUROC) of 0.88 and an area under the precision-recalled curve (AUPR) of 0.79. The novel approach proposed in this study suggests a new protocol for assessing cardiovascular diseases to act as a complementary tool to echocardiography as it provides insights on the circadian rhythm of the heart and can be potentially personalized according to patient clinical profile information.Clinical relevance- Implementing polar representations with deep learning in clinical settings to supplement echocardiography leverages continuous monitoring of the heart's circadian rhythm and personalized cardiovascular medicine while reducing the burden on medical practitioners.

The role of artificial intelligence in hypertensive disorders of pregnancy: towards personalized healthcare

INTRODUCTION

Guidelines advise ongoing follow-up of patients after hypertensive disorders of pregnancy (HDP) to assess cardiovascular risk and manage future patient-specific pregnancy conditions. However, there are limited tools available to monitor patients, with those available tending to be simple risk assessments that lack personalization. A promising approach could be the emerging artificial intelligence (AI)-based techniques, developed from big patient datasets to provide personalized recommendations for preventive advice.

AREAS COVERED

In this narrative review, we discuss the impact of integrating AI and big data analysis for personalized cardiovascular care, focusing on the management of HDP.

EXPERT OPINION

The pathophysiological response of women to pregnancy varies, and deeper insight into each response can be gained through a deeper analysis of the medical history of pregnant women based on clinical records and imaging data. Further research is required to be able to implement AI for clinical cases using multi-modality and multi-organ assessment, and this could expand both knowledge on pregnancy-related disorders and personalized treatment planning.

Dynamic Monitoring of Probiotics Effect in Parkinson's Disease Patients via Swarm Decomposition and Bispectral Analysis of Electrogastrograms

People with Parkinson's Disease (PwP) experience a significant deterioration of their daily life quality due to non-motor symptoms, with gastrointestinal dysfunctions manifesting as a vanguard of the latter. Electrogastrography (EGG) is a noninvasive diagnostic tool that can potentially provide biomarkers for the monitoring of dynamic gastric alterations that are related to daily lifestyle and treatment regimens. In this work, a robust analysis of EGG dynamics is introduced to evaluate the effect of probiotic treatment on PwP. The proposed framework, namely biSEGG, introduces a Swarm Decomposition-based enhancement of the EGG, combined with Bispectral feature engineering to model the underlying Quadratic Phase Coupling interactions between the gastric activity oscillatory components of EGG. The biSEGG features are benchmarked against the conventional Power Spectrum-based ones and evaluated through machine learning classifiers. The experimental results, when biSEGG was applied on data epochs from 11 PwP (probiotic vs placebo, AUROC: 0.67, Sensitivity/Specificity: 75/58%), indicate the superiority of biSEGG over Power Spectrum-based approaches and justify the efficiency of biSEGG in capturing and explaining intervention- and meal consumption-related alterations of the gastric activity in PwP.Clinical relevance- biSEGG holds potential for dynamic monitoring of gastrointestinal dysfunction and health status of PwP across diverse daily life scenarios.

Parkinsonian Tremor Detection with Compact Convolutional Transformer from Bispectrum Representation of tri-Axial Accelerometer Signals

After the breakthroughs of Transformer networks in Natural Language Processing (NLP) tasks, they have led to exciting progress in visual tasks as well. Nonetheless, there has been a parallel growth in the number of parameters and the amount of training data, which led to the conclusion that Transformers are not suited for small datasets. This paper is the first to convey the feasibility of Compact Convolutional Transformers (CCT) for the prediction of Parkinsonian postural tremor based on the Bispectrum (BS) representation of IMU accelerometer time series. The dataset includes tri-axial accelerometer signals collected unobtrusively in-the-wild while subjects are on a phone call, and labelled by neurologists and signal processing experts. The BS is a noise-immune, higher-order representation that reflects a signal's deviation from Gaussianity and measures quadratic phase coupling. We performed comparative classification experiments using the CCT, pre-trained CNNs such as VGG-16 and ResNet-50, and the conventional Vision Transformer (ViT). Our model achieves competitive prediction accuracy and F1 score of 96% with only 1.016 M trainable parameters, compared to the ViT with 21.659 M trainable parameters, in a five-fold cross-validation scheme. Our model also outperforms pre-trained CNNs such as VGG-16 and ResNet-50. Furthermore, we show that the performance gains are maintained when training on a larger dataset of BS images. Our effort here is motivated by the hypothesis that data-efficient transformers outperform transfer learning using pre-trained CNNs, paving the way for promising deep learning architecture for small-scale, novel and noisy medical imaging datasets.Clinical relevance- Novel deep learning model for unobtrusive prediction of Parkinsonian Postural Tremor from Bispectrum image representation of tri-axial accelerometer signals collected in-the-wild.

The convergence of traditional and digital biomarkers through AI-assisted biosensing: A new era in translational diagnostics?

Advances in consumer electronics, alongside the fields of microfluidics and nanotechnology have brought to the fore low-cost wearable/portable smart devices. Although numerous smart devices that track digital biomarkers have been successfully translated from bench-to-bedside, only a few follow the same fate when it comes to track traditional biomarkers. Current practices still involve laboratory-based tests, followed by blood collection, conducted in a clinical setting as they require trained personnel and specialized equipment. In fact, real-time, passive/active and robust sensing of physiological and behavioural data from patients that can feed artificial intelligence (AI)-based models can significantly improve decision-making, diagnosis and treatment at the point-of-procedure, by circumventing conventional methods of sampling, and in person investigation by expert pathologists, who are scarce in developing countries. This review brings together conventional and digital biomarker sensing through portable and autonomous miniaturized devices. We first summarise the technological advances in each field vs the current clinical practices and we conclude by merging the two worlds of traditional and digital biomarkers through AI/ML technologies to improve patient diagnosis and treatment. The fundamental role, limitations and prospects of AI in realizing this potential and enhancing the existing technologies to facilitate the development and clinical translation of "point-of-care" (POC) diagnostics is finally showcased.

Investigating the effects of beta-blockers on circadian heart rhythm using heart rate variability in ischemic heart disease with preserved ejection fraction

Heart failure is characterized by sympathetic activation and parasympathetic withdrawal leading to an abnormal autonomic modulation. Beta-blockers (BB) inhibit overstimulation of the sympathetic system and are indicated in heart failure patients with reduced ejection fraction. However, the effect of beta-blocker therapy on heart failure with preserved ejection fraction (HFpEF) is unclear. ECGs of 73 patients with HFpEF > 55% were recruited. There were 56 patients in the BB group and 17 patients in the without BB (NBB) group. The HRV analysis was performed for the 24-h period using a window size of 1,4 and 8-h. HRV measures between day and night for both the groups were also compared. Percentage change in the BB group relative to the NBB group was used as a measure of difference. RMSSD (13.27%), pNN50 (2.44%), HF power (44.25%) and LF power (13.53%) showed an increase in the BB group relative to the NBB group during the day and were statistically significant between the two groups for periods associated with high cardiac risk during the morning hours. LF:HF ratio showed a decrease of 3.59% during the day. The relative increase in vagal modulated RMSSD, pNN50 and HF power with a decrease in LF:HF ratio show an improvement in the parasympathetic tone and an overall decreased risk of a cardiac event especially during the morning hours that is characterized by a sympathetic surge.

A scoping review of neurodegenerative manifestations in explainable digital phenotyping

Neurologists nowadays no longer view neurodegenerative diseases, like Parkinson's and Alzheimer's disease, as single entities, but rather as a spectrum of multifaceted symptoms with heterogeneous progression courses and treatment responses. The definition of the naturalistic behavioral repertoire of early neurodegenerative manifestations is still elusive, impeding early diagnosis and intervention. Central to this view is the role of artificial intelligence (AI) in reinforcing the depth of phenotypic information, thereby supporting the paradigm shift to precision medicine and personalized healthcare. This suggestion advocates the definition of disease subtypes in a new biomarker-supported nosology framework, yet without empirical consensus on standardization, reliability and interpretability. Although the well-defined neurodegenerative processes, linked to a triad of motor and non-motor preclinical symptoms, are detected by clinical intuition, we undertake an unbiased data-driven approach to identify different patterns of neuropathology distribution based on the naturalistic behavior data inherent to populations in-the-wild. We appraise the role of remote technologies in the definition of digital phenotyping specific to brain-, body- and social-level neurodegenerative subtle symptoms, emphasizing inter- and intra-patient variability powered by deep learning. As such, the present review endeavors to exploit digital technologies and AI to create disease-specific phenotypic explanations, facilitating the understanding of neurodegenerative diseases as "bio-psycho-social" conditions. Not only does this translational effort within explainable digital phenotyping foster the understanding of disease-induced traits, but it also enhances diagnostic and, eventually, treatment personalization.

Users' Perspective on the AI-Based Smartphone PROTEIN App for Personalized Nutrition and Healthy Living: A Modified Technology Acceptance Model (mTAM) Approach

The ubiquitous nature of smartphone ownership, its broad application and usage, along with its interactive delivery of timely feedback are appealing for health-related behavior change interventions via mobile apps. However, users' perspectives about such apps are vital in better bridging the gap between their design intention and effective practical usage. In this vein, a modified technology acceptance model (mTAM) is proposed here, to explain the relationship between users' perspectives when using an AI-based smartphone app for personalized nutrition and healthy living, namely, PROTEIN, and the mTAM constructs toward behavior change in their nutrition and physical activity habits. In particular, online survey data from 85 users of the PROTEIN app within a period of 2 months were subjected to confirmatory factor analysis (CFA) and regression analysis (RA) to reveal the relationship of the mTAM constructs, i.e., perceived usefulness (PU), perceived ease of use (PEoU), perceived novelty (PN), perceived personalization (PP), usage attitude (UA), and usage intention (UI) with the users' behavior change (BC), as expressed via the acceptance/rejection of six related hypotheses (H1-H6), respectively. The resulted CFA-related parameters, i.e., factor loading (FL) with the related p-value, average variance extracted (AVE), and composite reliability (CR), along with the RA results, have shown that all hypotheses H1-H6 can be accepted (p < 0.001). In particular, it was found that, in all cases, FL > 0.5, CR > 0.7, AVE > 0.5, indicating that the items/constructs within the mTAM framework have good convergent validity. Moreover, the adjusted coefficient of determination (R 2) was found within the range of 0.224-0.732, justifying the positive effect of PU, PEoU, PN, and PP on the UA, that in turn positively affects the UI, leading to the BC. Additionally, using a hierarchical RA, a significant change in the prediction of BC from UA when the UI is used as a mediating variable was identified. The explored mTAM framework provides the means for explaining the role of each construct in the functionality of the PROTEIN app as a supportive tool for the users to improve their healthy living by adopting behavior change in their dietary and physical activity habits. The findings herein offer insights and references for formulating new strategies and policies to improve the collaboration among app designers, developers, behavior scientists, nutritionists, physical activity/exercise physiology experts, and marketing experts for app design/development toward behavior change.

Investigation of the Evolution of Wavelet Higher-Order Dynamics in Atrial Fibrillation

Atrial fibrillation (AF) is the most common cardiac arrhythmia and is associated with significant morbidity and mortality. Owing to the advances in sensor technology and the emergence of wearable devices that enable daily self-monitoring, ECG signal processing methods for the automatic detection of AF are more pertinent than ever. In this paper, we investigate the use of wavelet higher-order statistics (WHOS) for feature extraction and differentiation between normal sinus rhythm and AF. The proposed approach captures the evolution of the WHOS dynamics and quantifies the changes in the time-varying characteristics of the frequency couplings caused by AF. Results obtained from the statistical analysis of a dataset of 5834 single-lead ECG recordings, reveal 46/50 statistically significant features and provide insight into the complexity of the evolution of the ECG non-linearities during AF.

Diagnostic accuracy of keystroke dynamics as digital biomarkers for fine motor decline in neuropsychiatric disorders: a systematic review and meta-analysis

The unmet timely diagnosis requirements, that take place years after substantial neural loss and neuroperturbations in neuropsychiatric disorders, affirm the dire need for biomarkers with proven efficacy. In Parkinson's disease (PD), Mild Cognitive impairment (MCI), Alzheimers disease (AD) and psychiatric disorders, it is difficult to detect early symptoms given their mild nature. We hypothesize that employing fine motor patterns, derived from natural interactions with keyboards, also knwon as keystroke dynamics, could translate classic finger dexterity tests from clinics to populations in-the-wild for timely diagnosis, yet, further evidence is required to prove this efficiency. We have searched PubMED, Medline, IEEEXplore, EBSCO and Web of Science for eligible diagnostic accuracy studies employing keystroke dynamics as an index test for the detection of neuropsychiatric disorders as the main target condition. We evaluated the diagnostic performance of keystroke dynamics across 41 studies published between 2014 and March 2022, comprising 3791 PD patients, 254 MCI patients, and 374 psychiatric disease patients. Of these, 25 studies were included in univariate random-effect meta-analysis models for diagnostic performance assessment. Pooled sensitivity and specificity are 0.86 (95% Confidence Interval (CI) 0.82-0.90, I2 = 79.49%) and 0.83 (CI 0.79-0.87, I2 = 83.45%) for PD, 0.83 (95% CI 0.65-1.00, I2 = 79.10%) and 0.87 (95% CI 0.80-0.93, I2 = 0%) for psychomotor impairment, and 0.85 (95% CI 0.74-0.96, I2 = 50.39%) and 0.82 (95% CI 0.70-0.94, I2 = 87.73%) for MCI and early AD, respectively. Our subgroup analyses conveyed the diagnosis efficiency of keystroke dynamics for naturalistic self-reported data, and the promising performance of multimodal analysis of naturalistic behavioral data and deep learning methods in detecting disease-induced phenotypes. The meta-regression models showed the increase in diagnostic accuracy and fine motor impairment severity index with age and disease duration for PD and MCI. The risk of bias, based on the QUADAS-2 tool, is deemed low to moderate and overall, we rated the quality of evidence to be moderate. We conveyed the feasibility of keystroke dynamics as digital biomarkers for fine motor decline in naturalistic environments. Future work to evaluate their performance for longitudinal disease monitoring and therapeutic implications is yet to be performed. We eventually propose a partnership strategy based on a "co-creation" approach that stems from mechanistic explanations of patients' characteristics derived from data obtained in-clinics and under ecologically valid settings. The protocol of this systematic review and meta-analysis is registered in PROSPERO; identifier CRD42021278707. The presented work is supported by the KU-KAIST joint research center.

Machine Learning-Based Analysis of Digital Movement Assessment and ExerGame Scores for Parkinson's Disease Severity Estimation

Neurodegenerative Parkinson's Disease (PD) is one of the common incurable diseases among the elderly. Clinical assessments are characterized as standardized means for PD diagnosis. However, relying on medical evaluation of a patient's status can be subjective to physicians' experience, making the assessment process susceptible to human errors. The use of ICT-based tools for capturing the status of patients with PD can provide more objective and quantitative metrics. In this vein, the Personalized Serious Game Suite (PGS) and intelligent Motor Assessment Tests (iMAT), produced within the i-PROGNOSIS European project (www.i-prognosis.eu), are explored in the current study. More specifically, data from 27 patients with PD at Stage 1 (9) and Stage 3 (18) produced from their interaction with PGS/iMAT are analyzed. Five feature vector (FV) scenarios are set, including features from PGS or iMAT scores or their combination, after also taking into consideration the age of patients with PD. These FVs are fed into three machine learning classifiers, i.e., K-Nearest Neighbor (KNN), Support Vector Machines (SVM), and Random Forest (RF), to infer the stage of each patient with PD. A Leave-One-Out Cross-Validation (LOOCV) method is adopted for testing the classification performance. The experimental results show that a high (>90%) classification accuracy is achieved from both data sources (PGS/iMAT), justifying the effectiveness of PGS/iMAT to efficiently reflect the motor skill status of patients with PD and further potentiating PGS/iMAT enhancement with a machine learning a part to infer for the stage of patients with PD. Clearly, this integrated approach provides new opportunities for remote monitoring of the stage of patients with PD, contributing to a more efficient organization and set up of personalized interventions.

LSTM-Modeling of Emotion Recognition Using Peripheral Physiological Signals in Naturalistic Conversations

The automated recognition of human emotions plays an important role in developing machines with emotional intelligence. Major research efforts are dedicated to the development of emotion recognition methods. However, most of the affective computing models are based on images, audio, videos and brain signals. Literature lacks works that focus on utilizing only peripheral signals for emotion recognition (ER), which can be ideally implemented in daily life settings. Therefore, in this paper we present a framework for ER on the arousal and valence space, based on using multi-modal peripheral signals. The data used in this work were collected during a debate between two people using wearable devices. The emotions of the participants were rated by multiple raters and converted into classes in correspondence to the arousal and valence space. The use of a dynamic threshold for ratings conversion was investigated. An ER model is proposed that uses a Long Short-Term Memory (LSTM)-based architecture for classification. The model uses heart rate (HR), temperature (T), and electrodermal activity (EDA) signals as its inputs with emotional cues. Additionally, a post-processing prediction mechanism is introduced to enhance the recognition performance. The model is implemented to study the use of individual and different combinations of the peripheral signals, as well as utilizing annotations from different ratings. Additionally, it is employed for classification of valence and arousal in an independent and combined fashion, under subject dependent and independent scenarios. The experimental results have justified the efficient performance of the proposed framework, achieving classification accuracy 96% and 93% for the independent and combined classification scenarios, accordingly. The comparison of the achieved performance against the baseline methods shows the superiority of the proposed framework and the ability to recognize arousal-valance levels with high accuracy from peripheral signals, in real-life scenarios.

Deep Learning Predicts Heart Failure With Preserved, Mid-Range, and Reduced Left Ventricular Ejection Fraction From Patient Clinical Profiles

Background: Left ventricular ejection fraction (LVEF) is the gold standard for evaluating heart failure (HF) in coronary artery disease (CAD) patients. It is an essential metric in categorizing HF patients as preserved (HFpEF), mid-range (HFmEF), and reduced (HFrEF) ejection fraction but differs, depending on whether the ASE/EACVI or ESC guidelines are used to classify HF. Objectives: We sought to investigate the effectiveness of using deep learning as an automated tool to predict LVEF from patient clinical profiles using regression and classification trained models. We further investigate the effect of utilizing other LVEF-based thresholds to examine the discrimination ability of deep learning between HF categories grouped with narrower ranges. Methods: Data from 303 CAD patients were obtained from American and Greek patient databases and categorized based on the American Society of Echocardiography and the European Association of Cardiovascular Imaging (ASE/EACVI) guidelines into HFpEF (EF > 55%), HFmEF (50% ≤ EF ≤ 55%), and HFrEF (EF < 50%). Clinical profiles included 13 demographical and clinical markers grouped as cardiovascular risk factors, medication, and history. The most significant and important markers were determined using linear regression fitting and Chi-squared test combined with a novel dimensionality reduction algorithm based on arc radial visualization (ArcViz). Two deep learning-based models were then developed and trained using convolutional neural networks (CNN) to estimate LVEF levels from the clinical information and for classification into one of three LVEF-based HF categories. Results: A total of seven clinical markers were found important for discriminating between the three HF categories. Using statistical analysis, diabetes, diuretics medication, and prior myocardial infarction were found statistically significant (p < 0.001). Furthermore, age, body mass index (BMI), anti-arrhythmics medication, and previous ventricular tachycardia were found important after projections on the ArcViz convex hull with an average nearest centroid (NC) accuracy of 94%. The regression model estimated LVEF levels successfully with an overall accuracy of 90%, average root mean square error (RMSE) of 4.13, and correlation coefficient of 0.85. A significant improvement was then obtained with the classification model, which predicted HF categories with an accuracy ≥93%, sensitivity ≥89%, 1-specificity <5%, and average area under the receiver operating characteristics curve (AUROC) of 0.98. Conclusions: Our study suggests the potential of implementing deep learning-based models clinically to ensure faster, yet accurate, automatic prediction of HF based on the ASE/EACVI LVEF guidelines with only clinical profiles and corresponding information as input to the models. Invasive, expensive, and time-consuming clinical testing could thus be avoided, enabling reduced stress in patients and simpler triage for further intervention.

Swarm Decomposition of Abdominal Signals for Non-invasive Fetal ECG Extraction

The non-invasive fetal electrocardiography (fECG) extraction from maternal abdominal signals is one of the most promising modern fetal monitoring techniques. However, the noninvasive fECG signal is heavily contaminated with noise and overlaps with other prominent signals like the maternal ECG. In this work we propose a novel approach in non-invasive fECG extraction using the swarm decomposition (SWD) to isolate the fetal components from the abdominal signal. Accompanied with the use of higher-order statistics (HOS) for R peak detection, the application of the proposed method to the Abdominal and Direct Fetal ECG PhysioNet Database resulted in fetal R peak detection sensitivity of 99.8% and a positive predictability of 99.8%. Our results demonstrate the applicability of SWD and its potentiality in extracting fECG of good morphological quality with more deep decomposition levels, in order to connect the extracted structural characteristics of the fECG with the health status of the fetus.Clinical Relevance- The developed method shows improvement in fetal R peak detection for certain signals.

Detection of Mild Cognitive Impairment Through Natural Language and Touchscreen Typing Processing

Mild cognitive impairment (MCI), an identified prodromal stage of Alzheimer's Disease (AD), often evades detection in the early stages of the condition, when existing diagnostic methods are employed in the clinical setting. From an alternative perspective, smartphone interaction behavioral data, unobtrusively acquired in a non-clinical setting, can assist the screening and monitoring of MCI and its symptoms' progression. In this vein, the diagnostic ability of digital biomarkers, drawn from Fine Motor Impairment (FMI)- and Spontaneous Written Speech (SWS)-related data analysis, are examined here. In particular, keystroke dynamics derived from touchscreen typing activities, using Convolutional Neural Networks, along with linguistic features of SWS through Natural Language Processing (NLP), were used to distinguish amongst MCI patients and healthy controls (HC). Analytically, three indices of FMI (rigidity, bradykinesia and alternate finger tapping) and nine NLP features, related with lexical richness, grammatical, syntactical complexity, and word deficits, formed the feature space. The proposed approach was tested on two demographically matched groups of 11 MCI patients and 12 HC, having undergone the same neuropsychological tests, producing 4,930 typing sessions and 78 short texts, within 6 months, for analysis. A cascaded-classifier scheme was realized under three different feature combinations and validated via a Leave-One-Subject-Out cross-validation scheme. The acquired results have shown: (a) keystroke features with a k-NN classifier achieved an Area Under Curve (AUC) of 0.78 [95% confidence interval (CI):0.68-0.88; specificity/sensitivity (SP/SE): 0.64/0.92], (b) NLP features with a Logistic regression classifier achieved an AUC of 0.76 (95% CI: 0.65-0.85; SP/SE: 0.80/0.71), and (c) an ensemble model with the fusion of keystroke and NLP features resulted in AUC of 0.75 (95% CI:0.63-0.86; SP/SE 0.90/0.60). The current findings indicate the potentiality of new digital biomarkers to capture early stages of cognitive decline, providing a highly specific remote screening tool in-the-wild.

Parkinson's Disease Detection Based on Running Speech Data From Phone Calls

OBJECTIVE

Parkinson's Disease (PD) is a progressive neurodegenerative disorder, manifesting with subtle early signs, which often hinder timely and early diagnosis and treatment. The development of accessible, technology-based methods for longitudinal PD symptoms tracking in daily living offers the potential for transforming the disease assessment and accelerating PD diagnosis.

METHODS

A privacy-aware method for classifying PD patients and healthy controls (HC), on the grounds of speech impairment present in PD, is proposed here. Voice features from running speech signals were extracted from recordings passively captured over voice phone calls. Features are fed in a language-aware training of multiple- and single-instance learning classifiers, along with demographic variables, exploiting a multilingual cohort of 498 subjects (392/106 self-reported HC/PD patients) to classify PD.

RESULTS

By means of leave-one-subject-out cross-validation, the best-performing models yielded 0.69/0.68/0.63/0.83 area under the Receiver Operating Characteristic curve (AUC) for the binary classification of PD patient vs. HC in sub-cohorts of English/Greek/German/Portuguese-speaking subjects, respectively. Out-of-sample testing of the best performing models was conducted in an additional dataset, generated by 63 clinically-assessed subjects (24/39 HC/early PD patients). Testing has resulted in 0.84/0.93/0.83 AUC for the English/Greek/German-speaking sub-cohorts, respectively. Comparative analysis with other approaches for language-aware PD detection justified the efficiency of the proposed one, considering the ecological validity of the acquired voice data.

CONCLUSIONS

The present work demonstrates increased robustness in PD detection using voice data captured in-the-wild.

SIGNIFICANCE

A high-frequency, privacy-aware and unobtrusive PD screening tool is introduced for the first time, based on analysis of voice samples captured during routine phone calls.

DeepFoG: An IMU-Based Detection of Freezing of Gait Episodes in Parkinson's Disease Patients via Deep Learning

Freezing of Gait (FoG) is a movement disorder that mostly appears in the late stages of Parkinson’s Disease (PD). It causes incapability of walking, despite the PD patient’s intention, resulting in loss of coordination that increases the risk of falls and injuries and severely affects the PD patient’s quality of life. Stress, emotional stimulus, and multitasking have been encountered to be associated with the appearance of FoG episodes, while the patient’s functionality and self-confidence are constantly deteriorating. This study suggests a non-invasive method for detecting FoG episodes, by analyzing inertial measurement unit (IMU) data. Specifically, accelerometer and gyroscope data from 11 PD subjects, as captured from a single wrist-worn IMU sensor during continuous walking, are processed via Deep Learning for window-based detection of the FoG events. The proposed approach, namely DeepFoG, was evaluated in a Leave-One-Subject-Out (LOSO) cross-validation (CV) and 10-fold CV fashion schemes against its ability to correctly estimate the existence or not of a FoG episode at each data window. Experimental results have shown that DeepFoG performs satisfactorily, as it achieves 83%/88% and 86%/90% sensitivity/specificity, for LOSO CV and 10-fold CV schemes, respectively. The promising performance of the proposed DeepFoG reveals the potentiality of single-arm IMU-based real-time FoG detection that could guide effective interventions via stimuli, such as rhythmic auditory stimulation (RAS) and hand vibration. In this way, DeepFoG may scaffold the elimination of risk of falls in PD patients, sustaining their quality of life in everyday living activities.

Efficient syncope prediction from resting state clinical data using wavelet bispectrum and multilayer perceptron neural network

Neurally mediated syncope (NMS) is the most common type of syncope, and head up tilt test (HUTT) is, so far, the most appropriate tool to identify NMS. In this work, an effort to predict the NMS before performing the HUTT is attempted. To achieve this, the heart rate variability (HRV) at rest and during the first minutes of tilting position during HUTT was analyzed using both time and frequency domains. Various features from HRV regularity and complexity, along with wavelet higher-order spectrum (WHOS) analysis in low-frequency (LF) and high-frequency (HF) bands were examined. The experimental results from 26 patients with history of NMS have shown that at rest, a time domain entropy measure and WHOS-based features in LF band exhibit significant differences between positive and negative HUTT as well as among 10 healthy subjects and NMS patients. The best performance of multilayer perceptron neural network (MPNN) was achieved by using an input vector consisted of WHOS-based HRV features in the LF zone and systolic blood pressure from the resting period, yielding an accuracy of 89.7%, assessed by 5-fold cross-validation. The promising results presented here pave the way for an early prediction of the HUTT outcome from resting state, contributing to the identification of patients at higher risk NMS. The HRV analysis along with systolic blood pressure at rest predict NMS using a multilayer perceptron neural network.

Estimating Left Ventricle Ejection Fraction Levels Using Circadian Heart Rate Variability Features and Support Vector Regression Models

OBJECTIVES

The purpose of this study was to set an optimal fit of the estimated LVEF at hourly intervals from 24-hour ECG recordings and compare it with the fit based on two gold-standard guidelines.

METHODS

Support vector regression (SVR) models were applied to estimate LVEF from ECG derived heart rate variability (HRV) data in one-hour intervals from 24-hour ECG recordings of patients with either preserved, mid-range, or reduced LVEF, obtained from the Intercity Digital ECG Alliance (IDEAL) study. A step-wise feature selection approach was used to ensure the best possible estimations of LVEF levels.

RESULTS

The experimental results have shown that the lowest Root Mean Square Error (RMSE) between the original and estimated LVEF levels was during 3-4 am, 5-6 am and 6-7 pm.

CONCLUSION

The observations suggest these hours as possible times for intervention and optimal treatment outcomes. In addition, LVEF classifications following the ACCF/AHA guidelines leads to a more accurate assessment of mid-range LVEF.

SIGNIFICANCE

This study paves the way to explore the use of HRV features in the prediction of LVEF percentages as an indicator of disease progression, which may lead to an automated classification process for CAD patients.

Assistive HCI-Serious Games Co-design Insights: The Case Study of i-PROGNOSIS Personalized Game Suite for Parkinson's Disease

Human-Computer Interaction (HCI) and games set a new domain in understanding people's motivations in gaming, behavioral implications of game play, game adaptation to player preferences and needs for increased engaging experiences in the context of HCI serious games (HCI-SGs). When the latter relate with people's health status, they can become a part of their daily life as assistive health status monitoring/enhancement systems. Co-designing HCI-SGs can be seen as a combination of art and science that involves a meticulous collaborative process. The design elements in assistive HCI-SGs for Parkinson's Disease (PD) patients, in particular, are explored in the present work. Within this context, the Game-Based Learning (GBL) design framework is adopted here and its main game-design parameters are explored for the Exergames, Dietarygames, Emotional games, Handwriting games, and Voice games design, drawn from the PD-related i-PROGNOSIS Personalized Game Suite (PGS) (www.i-prognosis.eu) holistic approach. Two main data sources were involved in the study. In particular, the first one includes qualitative data from semi-structured interviews, involving 10 PD patients and four clinicians in the co-creation process of the game design, whereas the second one relates with data from an online questionnaire addressed by 104 participants spanning the whole related spectrum, i.e., PD patients, physicians, software/game developers. Linear regression analysis was employed to identify an adapted GBL framework with the most significant game-design parameters, which efficiently predict the transferability of the PGS beneficial effect to real-life, addressing functional PD symptoms. The findings of this work can assist HCI-SG designers for designing PD-related HCI-SGs, as the most significant game-design factors were identified, in terms of adding value to the role of HCI-SGs in increasing PD patients' quality of life, optimizing the interaction with personalized HCI-SGs and, hence, fostering a collaborative human-computer symbiosis.

Smartwatch-based Activity Analysis During Sleep for Early Parkinson's Disease Detection

Parkinson's Disease (PD) is the second most common neurodegenerative disorder with the non-motor symptoms preceding the motor impairment that is needed for clinical diagnosis. In the current study, an angle-based analysis that processes activity data during sleep from a smartwatch for quantification of sleep quality, when applied on controls and PD patients, is proposed. Initially, changes in their arm angle due to activity are captured from the smartwatch triaxial accelerometry data and used for the estimation of the corresponding binary state (awake/sleep). Then, sleep metrics (i.e., sleep efficiency index, total sleep time, sleep fragmentation index, sleep onset latency, and wake after sleep onset) are computed and used for the discrimination between controls and PD patients. A process of validation of the proposed approach when compared with the PSG-based ground truth in an in-the-clinic setting, resulted in comparable state estimation. Moreover, data from 15 early PD patients and 11 healthy controls were used as a test set, including 1,376 valid sleep recordings in-the-wild setting. The univariate analysis of the extracted sleep metrics achieved up to 0.77 AUC in early PD patients vs. healthy controls classification and exhibited a statistically significant correlation (up to 0.46) with the clinical PD Sleep Scale 2 counterpart Items. The findings of the proposed method show the potentiality to capture non-motor behavior from users' nocturnal activity to detect PD in the early stage.

Investigating Circadian Heart Rate Variability in Coronary Artery Disease Patients with Various Degrees of Left Ventricle Ejection Fraction

Early and noninvasive identification of heart failure progression is an important adjunct to successful and timely intervention. Severity of heart failure (HF) was assessed by Left Ventricular Ejection Fraction (LVEF). In this paper, we explore the circadian (24-hour) heart rate variability (HRV) features from ''normal" (EF >50%), "at-risk" (EF <40%), and "border-line" (40% ≤ EF ≤ 50%) patient data to determine whether HRV features can predict the stage of heart failure. All coronary artery disease (CAD) 24-hour circadian heart rate data were fitted by a cosinor analysis algorithm. Hourly HRV features from time- and frequency-domains were then extracted from all 24-hour patient data. A one-way ANOVA test was performed followed by a Tukey post-hoc multiple comparison test to investigate the differences among the three groups. The results showed a statistically significant difference between the three groups when using the normalized high frequency (HF Norm), low frequency peak (LF Peak), and the normalized very-low frequency (VLF Norm) for the 05:00-06:00 and 18:00-19:00 time periods. These results highlight a possible link between the circadian variation of sympathetic and parasympathetic nervous system activity and LVEF for CAD patients. The results could be useful in differentiating the various degrees of LVEF by using only noninvasive HRV features derived over a 24-hour period.Clinical relevance- The proposed method could be clinically useful to estimate the extent of LVEF associated with the severity of heart failure by recording the circadian variation of the heart rate in CAD patients. However, further clinical trials on a larger cohort of patients and controls are required.

Early Parkinson's Disease Detection via Touchscreen Typing Analysis using Convolutional Neural Networks

Parkinson's Disease (PD) is the second most common neurodegenerative disorder worldwide, causing both motor and non-motor symptoms. In the early stages, symptoms are mild and patients may ignore their existence. As a result, they do not undergo any related clinical examination; hence delaying their PD diagnosis. In an effort to remedy such delay, analysis of data passively captured from user's interaction with consumer technologies has been recently explored towards remote screening of early PD motor signs. In the current study, a smartphone-based method analyzing subjects' finger interaction with the smartphone screen is developed for the quantification of fine-motor skills decline in early PD using Convolutional Neural Networks. Experimental results from the analysis of keystroke typing in-the-clinic data from 18 early PD patients and 15 healthy controls have shown a classification performance of 0.89 Area Under the Curve (AUC) with 0.79/0.79 sensitivity/specificity, respectively. Evaluation of the generalization ability of the proposed approach was made by its application on typing data arising from a separate self-reported cohort of 27 PD patients' and 84 healthy controls' daily usage with their personal smartphones (data in-the-wild), achieving 0.79 AUC with 0.74/0.78 sensitivity/specificity, respectively. The results show the potentiality of the proposed approach to process keystroke dynamics arising from users' natural typing activity to detect PD, which contributes to the development of digital tools for remote pathological symptom screening.

Fetal Cardiac Timing Events Estimation from Doppler Ultrasound Signal Cepstrum Analysis

Early diagnosis of the cardiac abnormalities during the pregnancy may reduce the risk of perinatal morbidity and mortality. Doppler ultrasound signals (DUS), which is commonly used for monitoring the fetal heart rate, can also be used for identifying the event timings of fetal cardiac valve motions. In this paper, we propose a non-invasive technique to identify the fetal cardiac timing events on the basis of analysis of fetal DUS (10 normal subjects and 6 abnormal subjects). We proposed using the cepstrum analysis which enabled the frequency contents of the Doppler signals to be linked to the opening and closing of the heart's valves (Aortic and mitral). The time intervals from R peak of fetal ECG to opening and closing of aortic valve were found to be 123.83±7.41 (msec) and 222.85±21.21 (msec), respectively. The time intervals from R peak to opening and closing of mitral valve were found to be 295.03±13.62 (msec) and 31.97±3.34 (msec), respectively. This feature is essential for diagnosing the disorder of cardiac rhythm, extremely used for diagnosis of heart abnormalities.

Fetal Cardiac Timing Events Estimation From Doppler Ultrasound Signals Using Swarm Decomposition

Perinatal morbidity and mortality can be reduced when any cardiac abnormalities during a pregnancy are diagnosed early. Doppler Ultrasound Signals (DUS) are often used to monitor the heart rate of a fetus and they can also be used to identify the timing events of fetal cardiac valve motions. This paper proposed a novel, non-invasive technique which can be used to identify the fetal cardiac timing events based upon the analysis of fetal DUS (based upon 66 normal subjects belonging to three differing age groups) which can later be used to estimate fetal cardiac intervals from a DUS signal. The foundation of this method is a novel decomposition method referred to as Swarm Decomposition (SWD) which makes it possible for the frequency contents of Doppler signals to be associated with cardiac valve motions. These motions include the opening (o) and closing (c) of Aortic (A) and Mitral (M) valves. When compared the SWD method results to the Empirical Mode Decomposition for the validation, the fetal cardiac timings were estimated successfully when isolating the constituent parts of analyzed DUS signals with reduced complexity compared to EMD method. Pulsed Doppler images are used in order to verify the estimated timings. Three fetal age groups were assessed in terms of their cardiac intervals: 16–29, 30–35, and 36–41 weeks. The time intervals (Systolic Time Interval, STI), (Isovolumic Relaxation Time, IRT), and (Pre-ejection Period, PEP) were found to change significantly (p < 0.05) across the three age groups. The evaluation of fetal cardiac performance can be enhanced, given that these findings can be leveraged as sensitive markers throughout the process.

Continuous wavelet transform and higher-order spectrum: combinatory potentialities in breath sound analysis and electroencephalogram-based pain characterization

The combination of the continuous wavelet transform (CWT) with a higher-order spectrum (HOS) merges two worlds into one that conveys information regarding the non-stationarity, non-Gaussianity and nonlinearity of the systems and/or signals under examination. In the current work, the third-order spectrum (TOS), which is used to detect the nonlinearity and deviation from Gaussianity of two types of biomedical signals, that is, wheezes and electroencephalogram (EEG), is combined with the CWT to offer a time-scale representation of the examined signals. As a result, a CWT/TOS field is formed and a time axis is introduced, creating a time-bifrequency domain, which provides a new means for wheeze nonlinear analysis and dynamic EEG-based pain characterization. A detailed description and examples are provided and discussed to showcase the combinatory potential of CWT/TOS in the field of advanced signal processing.This article is part of the theme issue 'Redundancy rules: the continuous wavelet transform comes of age'.

Touchscreen typing-pattern analysis for detecting fine motor skills decline in early-stage Parkinson's disease

Parkinson’s disease (PD) is a degenerative movement disorder causing progressive disability that severely affects patients’ quality of life. While early treatment can produce significant benefits for patients, the mildness of many early signs combined with the lack of accessible high-frequency monitoring tools may delay clinical diagnosis. To meet this need, user interaction data from consumer technologies have recently been exploited towards unsupervised screening for PD symptoms in daily life. Similarly, this work proposes a method for detecting fine motor skills decline in early PD patients via analysis of patterns emerging from finger interaction with touchscreen smartphones during natural typing. Our approach relies on low-/higher-order statistical features of keystrokes timing and pressure variables, computed from short typing sessions. Features are fed into a two-stage multi-model classification pipeline that reaches a decision on the subject’s status (PD patient/control) by gradually fusing prediction probabilities obtained for individual typing sessions and keystroke variables. This method achieved an AUC = 0.92 and 0.82/0.81 sensitivity/specificity (matched groups of 18 early PD patients/15 controls) with discriminant features plausibly correlating with clinical scores of relevant PD motor symptoms. These findings suggest an improvement over similar approaches, thereby constituting a further step towards unobtrusive early PD detection from routine activities.

Exergames for Parkinson's Disease Patients: How Participatory Design Led to Technology Adaptation

Parkinson's disease (PD) is a neurodegenerative disorder that affects more than 10 million people worldwide. Assistive technology and exergames come to play a beneficial role in positive mood and socialization improvement, overall quality of life and improved confidence with everyday functional activities. More and more Exergames inserts in the market but how many of that are fitting the patient's needs? How many of that took into consideration theirs's opinion. This study describes the Minimum Viable Product (MVP) model "Develop-measure-learn" circle in a co-creation way with the PD patients to develop and improve Exergames for them, and the tools that are needed to accomplish. The most important outcome of this procedure was the proposed development of more realistic games, giving the researchers the step of starting the investigation of 3D solutions.

Bullying incidences identification within an immersive environment using HD EEG-based analysis: A Swarm Decomposition and Deep Learning approach

Bullying is an everlasting phenomenon and the first, yet difficult, step towards the solution is its detection. Conventional approaches for bullying incidence identification include questionnaires, conversations and psychological tests. Here, unlike the conventional approaches, two experiments are proposed that involve visual stimuli with cases of bullying- and non-bullying- related ones, set within a 2D (simple video preview) and a Virtual Reality (VR) (immersive video preview) context. In both experimental settings, brain activity is recorded using high density (HD) (256 channels) electroencephalogram (EEG), and analyzed to identify the bullying stimuli type (bullying/non-bullying) and context (2D/VR). The proposed classification analysis uses a convolutional neural network (CNN), applying deep learning on the oscillatory modes (OCMs) embedded within the raw HD EEG data. The extraction of OCMs from the HD EEG data is achieved with swarm decomposition (SWD), which efficiently accounts for the non-stationarity and noise contamination of the raw HD EEG data. Experimental results from 17 subjects indicate that the new SWD/CNN approach achieves high discrimination accuracy (AUC = 0.987 between bullying/non-bullying stimuli type; AUC = 0.975, between bullying/non-bullying stimuli type and 2D/VR context), paving the way for better understanding of how brain's responses could act as indicators of bullying experience within immersive environments.

Detecting fetal heart sounds by means of Fractal Dimension analysis in the Wavelet domain

Phonocardiography is a low-cost technique for the detection of fetal heart sounds (FHS) that can extend clinical auscultation in mobile and home care setups. The work presented here examines the transferability of a Wavelet Transform (WT)-based method that combines also Fractal Dimension (FD) analysis, previously proposed as WT-FD for the cases of lung and bowel sound analysis [4], to the extraction of FHSs. The WT-FD method has been evaluated with 12 simulated FHS signals and has shown promising results in terms of accuracy and performance (89%) in identifying the location of heartbeat, even in cases of signals with additive noise up to (6dB). This robustness paves the way for WT-FD testing in real FHSs, recorded under clinical setting, clearly contributing to better evaluation of the fetal heart functionality.

A Comparative Study on Fetal Heart Rates Estimated from Fetal Phonography and Cardiotocography

The aim of this study is to investigate that fetal heart rates (fHR) extracted from fetal phonocardiography (fPCG) could convey similar information of fHR from cardiotocography (CTG). Four-channel fPCG sensors made of low cost (<$1) ceramic piezo vibration sensor within 3D-printed casings were used to collect abdominal phonogram signals from 20 pregnant mothers (>34 weeks of gestation). A novel multi-lag covariance matrix-based eigenvalue decomposition technique was used to separate maternal breathing, fetal heart sounds (fHS) and maternal heart sounds (mHS) from abdominal phonogram signals. Prior to the fHR estimation, the fPCG signals were denoised using a multi-resolution wavelet-based filter. The proposed source separation technique was first tested in separating sources from synthetically mixed signals and then on raw abdominal phonogram signals. fHR signals extracted from fPCG signals were validated using simultaneous recorded CTG-based fHR recordings.The experimental results have shown that the fHR derived from the acquired fPCG can be used to detect periods of acceleration and deceleration, which are critical indication of the fetus' well-being. Moreover, a comparative analysis demonstrated that fHRs from CTG and fPCG signals were in good agreement (Bland Altman plot has mean = -0.21 BPM and ±2 SD = ±3) with statistical significance (p < 0.001 and Spearman correlation coefficient ρ = 0.95). The study findings show that fHR estimated from fPCG could be a reliable substitute for fHR from the CTG, opening up the possibility of a low cost monitoring tool for fetal well-being.

Fetal Heart Sounds Detection Using Wavelet Transform and Fractal Dimension

Phonocardiography is a non-invasive technique for the detection of fetal heart sounds (fHSs). In this study, analysis of fetal phonocardiograph (fPCG) signals, in order to achieve fetal heartbeat segmentation, is proposed. The proposed approach (namely WT–FD) is a wavelet transform (WT)-based method that combines fractal dimension (FD) analysis in the WT domain for the extraction of fHSs from the underlying noise. Its adoption in this field stems from its successful use in the fields of lung and bowel sounds de-noising analysis. The efficiency of the WT–FD method in fHS extraction has been evaluated with 19 simulated fHS signals, created for the present study, with additive noise up to (3 dB), along with the simulated fPCGs database available at PhysioBank. Results have shown promising performance in the identification of the correct location and morphology of the fHSs, reaching an overall accuracy of 89% justifying the efficacy of the method. The WT–FD approach effectively extracts the fHS signals from the noisy background, paving the way for testing it in real fHSs and clearly contributing to better evaluation of the fetal heart functionality.

A Hybrid EMD-Kurtosis Method for Estimating Fetal Heart Rate from Continuous Doppler Signals

Monitoring of fetal heart rate (FHR) is an important measure of fetal wellbeing during the months of pregnancy. Previous works on estimating FHR variability from Doppler ultrasound (DUS) signal mainly through autocorrelation analysis showed low accuracy when compared with heart rate variability (HRV) computed from fetal electrocardiography (fECG). In this work, we proposed a method based on empirical mode decomposition (EMD) and the kurtosis statistics to estimate FHR and its variability from DUS. Comparison between estimated beat-to-beat intervals using the proposed method and the autocorrelation function (AF) with respect to RR intervals computed from fECG as the ground truth was done on DUS signals from 44 pregnant mothers in the early (20 cases) and late (24 cases) gestational weeks. The new EMD-kurtosis method showed significant lower error in estimating the number of beats in the early group (EMD-kurtosis: 2.2% vs. AF: 8.5%, p < 0.01, root mean squared error) and the late group (EMD-kurtosis: 2.9% vs. AF: 6.2%). The EMD-kurtosis method was also found to be better in estimating mean beat-to-beat with an average difference of 1.6 ms from true mean RR compared to 19.3 ms by using the AF method. However, the EMD-kurtosis performed worse than AF in estimating SNDD and RMSSD. The proposed EMD-kurtosis method is more robust than AF in low signal-to-noise ratio cases and can be used in a hybrid system to estimate beat-to-beat intervals from DUS. Further analysis to reduce the estimated beat-to-beat variability from the EMD-kurtosis method is needed.

Fuzzy logic-based risk of fall estimation using smartwatch data as a means to form an assistive feedback mechanism in everyday living activities

This Letter aims to create a fuzzy logic-based assistive prevention tool for falls, based on accessible sensory technology, such as smartwatch, resulting in monitoring of the risk factors of falls caused by orthostatic hypotension (OH); a drop in systolic blood pressure (DSBP) >20 mmHg due to postural changes. Epidemiological studies have shown that OH is a high risk factor for falls and has a strong impact in quality of life (QoL) of the elderly's, especially for some cases such as Parkinsonians. Based on smartwatch data, it is explored here how statistical features of heart rate variability (HRV) can lead to DSBP prediction and estimation of the risk of fall. In this vein, a pilot study was conducted in collaboration with five Greek Parkinson's Foundation patients and ten healthy volunteers. Taking into consideration, the estimated DSBP and additional statistics of the user's medical/behavioural history, a fuzzy logic inference system was developed, to estimate the instantaneous risk of fall. The latter is fed back to the user with a mechanism chosen by him/her (i.e. vibration and/or sound), to prevent a possible fall, and also sent to the attentive carers and/or healthcare professionals for a home-based monitoring beyond the clinic. The proposed approach paves the way for effective exploitation of the contribution of smartwatch data, such as HRV, in the sustain of QoL in everyday living activities.

Potential of hybrid adaptive filtering in inflammatory lesion detection from capsule endoscopy images

A new feature extraction technique for the detection of lesions created from mucosal inflammations in Crohn’s disease, based on wireless capsule endoscopy (WCE) images processing is presented here. More specifically, a novel filtering process, namely Hybrid Adaptive Filtering (HAF), was developed for efficient extraction of lesion-related structural/textural characteristics from WCE images, by employing Genetic Algorithms to the Curvelet-based representation of images. Additionally, Differential Lacunarity (DLac) analysis was applied for feature extraction from the HAF-filtered images. The resulted scheme, namely HAF-DLac, incorporates support vector machines for robust lesion recognition performance. For the training and testing of HAF-DLac, an 800-image database was used, acquired from 13 patients who undertook WCE examinations, where the abnormal cases were grouped into mild and severe, according to the severity of the depicted lesion, for a more extensive evaluation of the performance. Experimental results, along with comparison with other related efforts, have shown that the HAF-DLac approach evidently outperforms them in the field of WCE image analysis for automated lesion detection, providing higher classification results, up to 93.8% (accuracy), 95.2% (sensitivity), 92.4% (specificity) and 92.6% (precision). The promising performance of HAF-DLac paves the way for a complete computer-aided diagnosis system that could support physicians’ clinical practice.

Gender effect in human brain responses to bottom-up and top-down attention using the EEG 3D-Vector Field Tomography

The effect of gender in rapidly allocating attention to objects, features or locations, as reflected in brain activity, is examined in this study. A visual-attention task, consisting of bottom-up (visual pop-out) and top-down (visual search) conditions during stimuli of four triangles, i.e., a target and three distractors, was engaged. In pop-out condition, both color and orientation of the distractors differed from target, while in search condition they differed only in orientation. During the task, high-density EEG (256 channels) data were recorded and analyzed by means of behavioral, event-related potentials, i.e., the P300 component and brain source localization analysis using 3D-Vector Field Tomography (3D-VFT). Twenty subjects (half female; 32±4.7 years old) participated in the experiments, performing 60 trials for each condition. Behavioral analysis revealed that both female and male outperformed in the pop-out condition compared to the search one, with respect to accuracy and reaction time, whereas no gender-related statistical significant differences were found. Nevertheless, in the search condition, higher P300 amplitudes were detected for females compared to males (p <; 7 · 10(-3)). Moreover, the findings suggested that the maximum activation in females was located mainly in the left inferior frontal and superior temporal gyri, whereas in males it was found in the right inferior frontal and superior temporal gyri. Overall, the experimental results show that visual attention depends on contributions from different brain lateralization linked to gender, posing important implications in studying developmental disorders, characterized by gender differences.

Phase-amplitude cross-frequency coupling in EEG-derived cortical time series upon an auditory perception task

Recent evidence suggests that cross-frequency coupling (CFC) plays an essential role in multi-scale communication across the brain. The amplitude of the high frequency oscillations, responsible for local activity, is modulated by the phase of the lower frequency activity, in a task and region-relevant way. In this paper, we examine this phase-amplitude coupling in a two-tone oddball paradigm for the low frequency bands (delta, theta, alpha, and beta) and determine the most prominent CFCs. Data consisted of cortical time series, extracted by applying three-dimensional vector field tomography (3D-VFT) to high density (256 channels) electroencephalography (HD-EEG), and CFC analysis was based on the phase-amplitude coupling metric, namely PAC. Our findings suggest CFC spanning across all brain regions and low frequencies. Stronger coupling was observed in the delta band, that is closely linked to sensory processing. However, theta coupling was reinforced in the target tone response, revealing a task-dependent CFC and its role in brain networks communication.

Use of adaptive hybrid filtering process in Crohn's disease lesion detection from real capsule endoscopy videos

The aim of this Letter is to present a new capsule endoscopy (CE) image analysis scheme for the detection of small bowel ulcers that relate to Crohn's disease. More specifically, this scheme is based on: (i) a hybrid adaptive filtering (HAF) process, that utilises genetic algorithms to the curvelet-based representation of images for efficient extraction of the lesion-related morphological characteristics, (ii) differential lacunarity (DL) analysis for texture feature extraction from the HAF-filtered images and (iii) support vector machines for robust classification performance. For the training of the proposed scheme, namely HAF-DL, an 800-image database was used and the evaluation was based on ten 30-second long endoscopic videos. Experimental results, along with comparison with other related efforts, have shown that the HAF-DL approach evidently outperforms the latter in the field of CE image analysis for automated lesion detection, providing higher classification results. The promising performance of HAF-DL paves the way for a complete computer-aided diagnosis system that could support the physicians' clinical practice.

Swarm intelligence for detecting interesting events in crowded environments

This paper focuses on detecting and localizing anomalous events in videos of crowded scenes, i.e., divergences from a dominant pattern. Both motion and appearance information are considered, so as to robustly distinguish different kinds of anomalies, for a wide range of scenarios. A newly introduced concept based on swarm theory, histograms of oriented swarms (HOS), is applied to capture the dynamics of crowded environments. HOS, together with the well-known histograms of oriented gradients, are combined to build a descriptor that effectively characterizes each scene. These appearance and motion features are only extracted within spatiotemporal volumes of moving pixels to ensure robustness to local noise, increase accuracy in the detection of local, nondominant anomalies, and achieve a lower computational cost. Experiments on benchmark data sets containing various situations with human crowds, as well as on traffic data, led to results that surpassed the current state of the art (SoA), confirming the method's efficacy and generality. Finally, the experiments show that our approach achieves significantly higher accuracy, especially for pixel-level event detection compared to SoA methods, at a low computational cost.

EEG-Based Tonic Cold Pain Characterization Using Wavelet Higher Order Spectral Features

A novel approach in tonic cold pain characterization, based on electroencephalograph (EEG) data analysis using wavelet higher order spectral (WHOS) features, is presented here. The proposed WHOS-based feature space extends the relative power spectrum-based (phase blind) approaches reported so far a step forward; this is realized via dynamic monitoring of the nonlinerities of the EEG brain response to tonic cold pain stimuli by capturing the change in the underlying quadratic phase coupling at the bifrequency wavelet bispectrum/bicoherence domain due to the change of the pain level. Three pain characterization scenarios were formed and experimentally tested involving WHOS-based analysis of EEG data, acquired from 17 healthy volunteers that were subjected to trials of tonic cold pain stimuli. The experimental and classification analysis results, based on four well-known classifiers, have shown that the WHOS-based features successfully discriminate relax from pain status, provide efficient identification of the transition from relax to mild and/or severe pain status, and translate the subjective perception of pain to an objective measure of pain endurance. These findings seem quite promising and pave the way for adopting WHOS-based approaches to pain characterization under other types of pain, e.g., chronic pain and various clinical scenarios.