RESUMO
INTRODUCTION: This study examined individual and conjoint factors associated with beliefs about the harmfulness of nicotine replacement therapies (NRTs) relative to combustible cigarettes (CCs). AIMS AND METHODS: Data analyzed came from 8642 adults (≥18 years) who smoked daily/weekly and participated in the 2020 ITC Four Country Smoking and Vaping Survey in Australia (n = 1213), Canada (n = 2633), England (n = 3057), and United States (n = 1739). Respondents were asked: "Compared to smoking cigarettes, how harmful do you think nicotine replacement products are?" Responses were dichotomized into "much less" versus otherwise for analysis using multivariable logistic regression models, complemented by decision-tree analysis to identify conjoint factors. RESULTS: Percentages believing that NRTs are much less harmful than CCs were 29.7% (95% CI = 26.2% to 33.5%) in Australia, 27.4% (95% CI = 25.1% to 29.8%) in England, 26.4% (95% CI = 24.4% to 28.4%) in Canada, and 21.7% (95% CI = 19.2% to 24.3%) in the United States. Across all countries, believing nicotine is not at all/slightly harmful to health (aOR = 1.53-2.27), endorsing nicotine vaping products (NVPs) as less harmful than CCs (much less harmful: aOR = 7.24-14.27; somewhat less harmful: aOR = 1.97-3.23), and possessing higher knowledge of smoking harms (aOR = 1.23-1.88) were individual factors associated with increased odds of believing NRTs are much less harmful than CCs. With some country variations, these nicotine-related measures also interacted with each other and sociodemographic variables to serve as conjoint factors associated with the likelihood of accurate NRT relative harm belief. CONCLUSIONS: Many people who regularly smoke cigarettes are unaware that NRTs are much less harmful than cigarettes. Additionally, beliefs about NRTs relative harmfulness appear to be influenced by both individual and conjoint factors. IMPLICATIONS: This study demonstrates that despite past efforts to educate people who smoke about the harms of NRTs relative to CCs, misperceptions around the relative harmfulness of NRTs remain substantial. In all four studied countries, subgroups of people who smoke regularly who are misinformed about the relative harmfulness of NRTs, and who may be reluctant to use NRTs for smoking cessation can be reliably identified for corrective interventions based on their understanding of the harms related to nicotine, NVPs and smoking along with sociodemographic markers. The identified subgroup information can be used to prioritize and inform the development of effective interventions to specifically address the gaps in knowledge and understanding of the various subgroups identified. Our results suggest these may need to be tailored for each country.
Assuntos
Sistemas Eletrônicos de Liberação de Nicotina , Abandono do Hábito de Fumar , Produtos do Tabaco , Vaping , Adulto , Humanos , Estados Unidos/epidemiologia , Nicotina/efeitos adversos , Vaping/efeitos adversos , Dispositivos para o Abandono do Uso de Tabaco , Produtos do Tabaco/efeitos adversos , Inquéritos e QuestionáriosRESUMO
Heart sound signals reflect valuable information about heart condition. Previous studies have suggested that the information contained in single-channel heart sound signals can be used to detect coronary artery disease (CAD). But accuracy based on single-channel heart sound signal is not satisfactory. This paper proposed a method based on multi-domain feature fusion of multi-channel heart sound signals, in which entropy features and cross entropy features are also included. A total of 36 subjects enrolled in the data collection, including 21 CAD patients and 15 non-CAD subjects. For each subject, five-channel heart sound signals were recorded synchronously for 5 min. After data segmentation and quality evaluation, 553 samples were left in the CAD group and 438 samples in the non-CAD group. The time-domain, frequency-domain, entropy, and cross entropy features were extracted. After feature selection, the optimal feature set was fed into the support vector machine for classification. The results showed that from single-channel to multi-channel, the classification accuracy has increased from 78.75% to 86.70%. After adding entropy features and cross entropy features, the classification accuracy continued to increase to 90.92%. The study indicated that the method based on multi-domain feature fusion of multi-channel heart sound signals could provide more information for CAD detection, and entropy features and cross entropy features played an important role in it.
RESUMO
Entropy profiling is a recently introduced approach that reduces parametric dependence in traditional Kolmogorov-Sinai (KS) entropy measurement algorithms. The choice of the threshold parameter r of vector distances in traditional entropy computations is crucial in deciding the accuracy of signal irregularity information retrieved by these methods. In addition to making parametric choices completely data-driven, entropy profiling generates a complete profile of entropy information as against a single entropy estimate (seen in traditional algorithms). The benefits of using "profiling" instead of "estimation" are: (a) precursory methods such as approximate and sample entropy that have had the limitation of handling short-term signals (less than 1000 samples) are now made capable of the same; (b) the entropy measure can capture complexity information from short and long-term signals without multi-scaling; and (c) this new approach facilitates enhanced information retrieval from short-term HRV signals. The novel concept of entropy profiling has greatly equipped traditional algorithms to overcome existing limitations and broaden applicability in the field of short-term signal analysis. In this work, we present a review of KS-entropy methods and their limitations in the context of short-term heart rate variability analysis and elucidate the benefits of using entropy profiling as an alternative for the same.
RESUMO
QT interval variability (QTV) and heart rate variability (HRV) are both accepted biomarkers for cardiovascular events. QTV characterizes the variations in ventricular depolarization and repolarization. It is a predominant element of HRV. However, QTV is also believed to accept direct inputs from upstream control system. How QTV varies along with HRV is yet to be elucidated. We studied the dynamic relationship of QTV and HRV during different physiological conditions from resting, to cycling, and to recovering. We applied several entropy-based measures to examine their bivariate relationships, including cross sample entropy (XSampEn), cross fuzzy entropy (XFuzzyEn), cross conditional entropy (XCE), and joint distribution entropy (JDistEn). Results showed no statistically significant differences in XSampEn, XFuzzyEn, and XCE across different physiological states. Interestingly, JDistEn demonstrated significant decreases during cycling as compared with that during the resting state. Besides, JDistEn also showed a progressively recovering trend from cycling to the first 3 min during recovering, and further to the second 3 min during recovering. It appeared to be fully recovered to its level in the resting state during the second 3 min during the recovering phase. The results suggest that there is certain nonlinear temporal relationship between QTV and HRV, and that the JDistEn could help unravel this nuanced property.
RESUMO
The complexity of a heart rate variability (HRV) signal is considered an important nonlinear feature to detect cardiac abnormalities. This work aims at explaining the physiological meaning of a recently developed complexity measurement method, namely, distribution entropy (DistEn), in the context of HRV signal analysis. We thereby propose modified distribution entropy (mDistEn) to remove the physiological discrepancy involved in the computation of DistEn. The proposed method generates a distance matrix that is devoid of over-exerted multi-lag signal changes. Restricted element selection in the distance matrix makes "mDistEn" a computationally inexpensive and physiologically more relevant complexity measure in comparison to DistEn.
RESUMO
OBJECTIVE: To investigate the characteristics of motor manifestation during convulsive epileptic and psychogenic nonepileptic seizures (PNES), captured using a wrist-worn accelerometer (ACM) device. The main goal was to find quantitative ACM features that can differentiate between convulsive epileptic and convulsive PNES. METHODS: In this study, motor data were recorded using wrist-worn ACM-based devices. A total of 83 clinical events were recorded: 39 generalized tonic-clonic seizures (GTCS) from 12 patients with epilepsy, and 44 convulsive PNES from 7 patients (one patient had both GTCS and PNES). The temporal variations in the ACM traces corresponding to 39 GTCS and 44 convulsive PNES events were extracted using Poincaré maps. Two new indices-tonic index (TI) and dispersion decay index (DDI)-were used to quantify the Poincaré-derived temporal variations for every GTCS and convulsive PNES event. RESULTS: The TI and DDI of Poincaré-derived temporal variations for GTCS events were higher in comparison to convulsive PNES events (P < 0.001). The onset and the subsiding patterns captured by TI and DDI differentiated between epileptic and convulsive nonepileptic seizures. An automated classifier built using TI and DDI of Poincaré-derived temporal variations could correctly differentiate 42 (sensitivity: 95.45%) of 44 convulsive PNES events and 37 (specificity: 94.87%) of 39 GTCS events. A blinded review of the Poincaré-derived temporal variations in GTCS and convulsive PNES by epileptologists differentiated 26 (sensitivity: 70.27%) of 44 PNES events and 33 (specificity: 86.84%) of 39 GTCS events correctly. SIGNIFICANCE: In addition to quantifying the motor manifestation mechanism of GTCS and convulsive PNES, the proposed approach also has diagnostic significance. The new ACM features incorporate clinical characteristics of GTCS and PNES, thus providing an accurate, low-cost, and practical alternative to differential diagnosis of PNES.
Assuntos
Acelerometria/métodos , Epilepsia/diagnóstico , Movimento/fisiologia , Convulsões/diagnóstico , Dispositivos Eletrônicos Vestíveis/tendências , Punho/fisiologia , Adulto , Diagnóstico Diferencial , Eletroencefalografia/métodos , Epilepsia/fisiopatologia , Feminino , Humanos , Masculino , Convulsões/fisiopatologia , Fatores de Tempo , Adulto JovemRESUMO
OBJECTIVE: It is unclear whether blockade of the angiotensin system has effects on mental health. Our objective was to determine the impact of angiotensin converting enzyme inhibitors and angiotensin II type 1 receptor (AT1R) blockers on mental health domain of quality of life. STUDY DESIGN: Meta-analysis of published literature. DATA SOURCES: PubMed and clinicaltrials.gov databases. The last search was conducted in January 2017. STUDY SELECTION: Randomized controlled trials comparing any angiotensin converting enzyme inhibitor or AT1R blocker versus placebo or non-angiotensin converting enzyme inhibitor or non-AT1R blocker were selected. Study participants were adults without any major physical symptoms. We adhered to meta-analysis reporting methods as per PRISMA and the Cochrane Collaboration. DATA SYNTHESIS: Eleven studies were included in the analysis. When compared with placebo or other antihypertensive medications, AT1R blockers and angiotensin converting enzyme inhibitors were associated with improved overall quality of life (standard mean difference = 0.11, 95% confidence interval = [0.08, 0.14], p < 0.0001), positive wellbeing (standard mean difference = 0.11, 95% confidence interval = [0.05, 0.17], p < 0.0001), mental (standard mean difference = 0.15, 95% confidence interval = [0.06, 0.25], p < 0.0001), and anxiety (standard mean difference = 0.08, 95% confidence interval = [0.01, 0.16], p < 0.0001) domains of QoL. No significant difference was found for the depression domain (standard mean difference = 0.05, 95% confidence interval = [0.02, 0.12], p = 0.15). CONCLUSIONS: Use of angiotensin blockers and inhibitors for the treatment of hypertension in otherwise healthy adults is associated with improved mental health domains of quality of life. Mental health quality of life was a secondary outcome in the included studies. Research specifically designed to analyse the usefulness of drugs that block the angiotensin system is necessary to properly evaluate this novel psychiatric target.
Assuntos
Bloqueadores do Receptor Tipo 1 de Angiotensina II/farmacologia , Inibidores da Enzima Conversora de Angiotensina/farmacologia , Ansiedade/terapia , Hipertensão/tratamento farmacológico , Saúde Mental , Qualidade de Vida , Ensaios Clínicos Controlados Aleatórios como Assunto , HumanosRESUMO
BACKGROUND: Heart rate fluctuates beat-by-beat asymmetrically which is known as heart rate asymmetry (HRA). It is challenging to assess HRA robustly based on short-term heartbeat interval series. METHOD: An area index (AI) was developed that combines the distance and phase angle information of points in the Poincaré plot. To test its performance, the AI was used to classify subjects with: (i) arrhythmia, and (ii) congestive heart failure, from the corresponding healthy controls. For comparison, the existing Porta's index (PI), Guzik's index (GI), and slope index (SI) were calculated. To test the effect of data length, we performed the analyses separately using long-term heartbeat interval series (derived from >3.6-h ECG) and short-term segments (with length of 500 intervals). A second short-term analysis was further carried out on series extracted from 5-min ECG. RESULTS: For long-term data, SI showed acceptable performance for both tasks, i.e., for task i p < 0.001, Cohen's d = 0.93, AUC (area under the receiver-operating characteristic curve) = 0.86; for task ii p < 0.001, d = 0.88, AUC = 0.75. AI performed well for task ii (p < 0.001, d = 1.0, AUC = 0.78); for task i, though the difference was statistically significant (p < 0.001, AUC = 0.76), the effect size was small (d = 0.11). PI and GI failed in both tasks (p > 0.05, d < 0.4, AUC < 0.7 for all). However, for short-term segments, AI indicated better distinguishability for both tasks, i.e., for task i, p < 0.001, d = 0.71, AUC = 0.71; for task ii, p < 0.001, d = 0.93, AUC = 0.74. The rest three measures all failed with small effect sizes and AUC values (d < 0.5, AUC < 0.7 for all) although the difference in SI for task i was statistically significant (p < 0.001). Besides, AI displayed smaller variations across different short-term segments, indicating more robust performance. Results from the second short-term analysis were in keeping with those findings. CONCLUSION: The proposed AI indicated better performance especially for short-term heartbeat interval data, suggesting potential in the ambulatory application of cardiovascular monitoring.
Assuntos
Eletrocardiografia , Frequência Cardíaca , Processamento de Sinais Assistido por Computador , Feminino , Humanos , Masculino , Pessoa de Meia-IdadeRESUMO
BACKGROUND: As more and more researchers are turning to big data for new opportunities of biomedical discoveries, machine learning models, as the backbone of big data analysis, are mentioned more often in biomedical journals. However, owing to the inherent complexity of machine learning methods, they are prone to misuse. Because of the flexibility in specifying machine learning models, the results are often insufficiently reported in research articles, hindering reliable assessment of model validity and consistent interpretation of model outputs. OBJECTIVE: To attain a set of guidelines on the use of machine learning predictive models within clinical settings to make sure the models are correctly applied and sufficiently reported so that true discoveries can be distinguished from random coincidence. METHODS: A multidisciplinary panel of machine learning experts, clinicians, and traditional statisticians were interviewed, using an iterative process in accordance with the Delphi method. RESULTS: The process produced a set of guidelines that consists of (1) a list of reporting items to be included in a research article and (2) a set of practical sequential steps for developing predictive models. CONCLUSIONS: A set of guidelines was generated to enable correct application of machine learning models and consistent reporting of model specifications and results in biomedical research. We believe that such guidelines will accelerate the adoption of big data analysis, particularly with machine learning methods, in the biomedical research community.
Assuntos
Pesquisa Biomédica/métodos , Interpretação Estatística de Dados , Aprendizado de Máquina , Pesquisa Biomédica/normas , Humanos , Estudos Interdisciplinares , Modelos BiológicosRESUMO
Epilepsy is a life-threatening neurological condition. Manual detection of epileptic seizures (ES) is laborious and burdensome. Machine learning techniques applied to electroencephalography (EEG) signals are widely used for automatic seizure detection. Some key factors are worth considering for the real-world applicability of such systems: (i) continuous EEG data typically has a higher class imbalance; (ii) higher variability across subjects is present in physiological signals such as EEG; and (iii) seizure event detection is more practical than random segment detection. Most prior studies failed to address these crucial factors altogether for seizure detection. In this study, we intend to investigate a generalized cross-subject seizure event detection system using the continuous EEG signals from the CHB-MIT dataset that considers all these overlooked aspects. A 5-second non-overlapping window is used to extract 92 features from 22 EEG channels; however, the most significant 32 features from each channel are used in experimentation. Seizure classification is done using a Random Forest (RF) classifier for segment detection, followed by a post-processing method used for event detection. Adopting all the above-mentioned essential aspects, the proposed event detection system achieved 72.63% and 75.34% sensitivity for subject-wise 5-fold and leave-one-out analyses, respectively. This study presents the real-world scenario for ES event detectors and furthers the understanding of such detection systems.
RESUMO
Major depressive disorder (MDD) is a chronic mental illness which affects people's well-being and is often detected at a later stage of depression with a likelihood of suicidal ideation. Early detection of MDD is thus necessary to reduce the impact, however, it requires monitoring vitals in daily living conditions. EEG is generally multi-channel and due to difficulty in signal acquisition, it is unsuitable for home-based monitoring, whereas, wearable sensors can collect single-channel ECG. Classical machine-learning based MDD detection studies commonly use various heart rate variability features. Feature generation, which requires domain knowledge, is often challenging, and requires computation power, often unsuitable for real time processing, MDDBranchNet is a proposed parallel-branch deep learning model for MDD binary classification from a single channel ECG which uses additional ECG-derived signals such as R-R signal and degree distribution time series of horizontal visibility graph. The use of derived branches was able to increase the model's accuracy by around 7%. An optimal 20-second overlapped segmentation of ECG recording was found to be beneficial with a 70% prediction threshold for maximum MDD detection with a minimum false positive rate. The proposed model evaluated MDD prediction from signal excerpts, irrespective of location (first, middle or last one-third of the recording), instead of considering the entire ECG signal with minimal performance variation stressing the idea that MDD phenomena are likely to manifest uniformly throughout the recording.
Assuntos
Aprendizado Profundo , Transtorno Depressivo Maior , Eletrocardiografia , Processamento de Sinais Assistido por Computador , Humanos , Eletrocardiografia/métodos , Transtorno Depressivo Maior/fisiopatologia , Transtorno Depressivo Maior/diagnóstico , Algoritmos , Adulto , MasculinoRESUMO
BACKGROUND AND OBJECTIVE: Automatic electrocardiogram (ECG) signal analysis for heart disease detection has gained significant attention due to busy lifestyles. However, ECG signals are susceptible to noise, which adversely affects the performance of ECG signal analysers. Traditional blind filtering methods use predefined noise frequency and filter order, but they alter ECG biomarkers. Several Deep Learning-based ECG noise detection and classification methods exist, but no study compares recurrent neural network (RNN) and convolutional neural network (CNN) architectures and their complexity. METHODS: This paper introduces a knowledge-based ECG filtering system using Deep Learning to classify ECG noise types and compare popular computer vision model architectures in a practical Internet of Medical Things (IoMT) framework. Experimental results demonstrate that the CNN-based ECG noise classifier outperforms the RNN-based model in terms of performance and training time. RESULTS: The study shows that AlexNet, visual geometry group (VGG), and residual network (ResNet) achieved over 70% accuracy, specificity, sensitivity, and F1 score across six datasets. VGG and ResNet performances were comparable, but VGG was more complex than ResNet, with only a 4.57% less F1 score. CONCLUSIONS: This paper introduces a Deep Learning (DL) based ECG noise classifier for a knowledge-driven ECG filtering system, offering selective filtering to reduce signal distortion. Evaluation of various CNN and RNN-based models reveals VGG and Resnet outperform. Further, the VGG model is superior in terms of performance. But Resnet performs comparably to VGG with less model complexity.
Assuntos
Aprendizado Profundo , Eletrocardiografia , Redes Neurais de Computação , Processamento de Sinais Assistido por Computador , Eletrocardiografia/métodos , Humanos , Algoritmos , Razão Sinal-RuídoRESUMO
Recent studies have shown the potential of wearable sensors for objective detection of health and safety risks in construction workers through their collected physiological data. Body temperature, as the focus of the current study, is one of the most important physiological parameters that can help to detect various health and safety risks such as heat stress, physical fatigue, and infectious diseases. This study aims to assess the applicability and performance of off-the-shelf wearable sensor devices to monitor workers' body temperature in construction sites by evaluating the accuracy of temperature measurements as well as the comfort of the devices. A total of nine off-the-shelf wearable sensor devices available on the market were initially trialed in the laboratory, and three devices were shortlisted considering a set of selection criteria for further assessment. Over three weeks, the shortlisted wearable sensors were tested on 26 workers in two large construction sites in Australia. The reliability/validity of the selected wearable sensors in measuring body temperature was investigated using Bland-Altman analysis. Human factors were also investigated in terms of the comfort of the devices, their impact on workers' performance, and the acceptability of being worn for an extended period (i.e., 8 h or more). It was found that all selected devices measured body temperature with a bias of less than one indicating a slight difference in measurements compared to the reference hospital-grade thermometers. Two devices out of the three were also comfortable. The achieved results indicate that it is feasible to develop a continuous temperature monitoring platform using off-the-shelf wearable sensors to detect a range of significant health and safety risks in construction sites objectively. Considering the rapid advancements in manufacturing wearable sensors, future research can adopt a similar approach to include the newly introduced off-the-shelf temperature sensors and select the most appropriate device.
RESUMO
Interpretability often seeks domain-specific facts, which is understandable to human, from deep-learning (DL) or other machine-learning (ML) models of black-box nature. This is particularly important to establish transparency in ML model's inner-working and decision-making, so that a certain level of trust is achieved when a model is deployed in a sensitive and mission-critical context, such as health-care. Model-level transparency can be achieved when its components are transparent and are capable of explaining reason of a decision, for a given input, which can be linked to domain-knowledge. This article used convolutional neural network (CNN), with sinc-convolution as its constrained first-layer, to explore if such a model's decision-making can be explained, for a given task, by observing the sinc-convolution's sinc-kernels. These kernels work like band-pass filters, having only two parameters per kernel - lower and upper cutoff frequencies, and optimised through back-propagation. The optimised frequency-bands of sinc-kernels may provide domain-specific insights for a given task. For a given input instance, the effects of sinc-kernels was visualised by means of explanation vector, which may help to identify comparatively significant frequency-bands, that may provide domain-specific interpretation, for the given task. In addition, a CNN model was further optimised by considering the identified subset of prominent sinc frequency-bands as the constrained first-layer, which yielded comparable or better performance, as compared to its all sinc-bands counterpart, as well as, a classical CNN. A minimal CNN structure, achieved through such an optimisation process, may help design task-specific interpretable models. To the best of our knowledge, the idea of sinc-convolution layer's task-specific significant sinc-kernel-based network optimisation is the first of its kind. Additionally, the idea of explanation-vector-based joint time-frequency representation to analyse time-series signals is rare in the literature. The above concept was validated for two tasks, ECG beat-classification (five-class classification task), and R-peak localisation (sample-wise segmentation task).
Assuntos
Algoritmos , Redes Neurais de Computação , Humanos , Aprendizado de MáquinaRESUMO
Deep-learning-based QRS-detection algorithms often require essential post-processing to refine the output prediction-stream for R-peak localisation. The post-processing involves basic signal-processing tasks including the removal of random noise in the model's prediction stream using a basic Salt and Pepper filter, as well as, tasks that use domain-specific thresholds, including a minimum QRS size, and a minimum or maximum R-R distance. These thresholds were found to vary among QRS-detection studies and empirically determined for the target dataset, which may have implications if the target dataset differs such as the drop of performance in unknown test datasets. Moreover, these studies, in general, fail to identify the relative strengths of deep-learning models and the post-processing to weigh them appropriately. This study identifies the domain-specific post-processing, as found in the QRS-detection literature, as three steps based on the required domain knowledge. It was found that the use of minimal domain-specific post-processing is often sufficient for most of the cases and the use of additional domain-specific refinement ensures superior performance, however, it makes the process biased towards the training data and lacks generalisability. As a remedy, a domain-agnostic automated post-processing is introduced where a separate recurrent neural network (RNN)-based model learns required post-processing from the output generated from a QRS-segmenting deep learning model, which is, to the best of our knowledge, the first of its kind. The RNN-based post-processing shows superiority over the domain-specific post-processing for most of the cases (with shallow variants of the QRS-segmenting model and datasets like TWADB) and lags behind for others but with a small margin ( ≤ 2%). The consistency of the RNN-based post-processor is an important characteristic which can be utilised in designing a stable and domain agnostic QRS detector.
Assuntos
Algoritmos , Redes Neurais de Computação , Processamento de Sinais Assistido por ComputadorRESUMO
Continuous monitoring of stress in individuals during their daily activities has become an inevitable need in present times. Unattended stress is a silent killer and may lead to fatal physical and mental disorders if left unidentified. Stress identification based on individual judgement often leads to under-diagnosis and delayed treatment possibilities. EEG-based stress monitoring is quite popular in this context, but impractical to use for continuous remote monitoring.Continuous remote monitoring of stress using signals acquired from everyday wearables like smart watches is the best alternative here. Non-EEG data such as heart rate and ectodermal activity can also act as indicators of physiological stress. In this work, we have explored the possibility of using nonlinear features from non-EEG data such as (a) heart rate, (b) ectodermal activity, (c) body temperature (d) SpO2 and (e) acceleration in detecting four different types of neurological states; namely (1) Relaxed state, (2) State of Physical stress, (3) State of Cognitive stress and (4) State of Emotional stress. Physiological data of 20 healthy adults have been used from the noneeg database of PhysioNet.Results: We used two machine learning models; a linear logistic regression and a nonlinear random forest to detect (a) stress from relaxed state and (4) the four different neurological states. We trained the models using linear and nonlinear features separately. For the 2-class and 4-class problems, using nonlinear features increased the accuracy of the models. Moreover, it is also proved in this study that by using nonlinear features, we can avoid the use of complex machine learning models.
Assuntos
Eletroencefalografia , Transtornos Mentais , Adulto , Humanos , Vigília , Aprendizado de Máquina , Frequência CardíacaRESUMO
Automatic signal analysis using artificial intelligence is getting popular in digital healthcare, such as ECG rhythm analysis, where ECG signals are collected from traditional ECG machines or wearable ECG sensors. However, the risk of using an automated system for ECG analysis when noise is present can lead to incorrect diagnosis or treatment decisions. A noise detector is crucial to minimise the risk of incorrect diagnosis. Machine learning (ML) models are used in ECG noise detection before clinical decision-making systems to mitigate false alarms. However, it is essential to prove the generalisation capability of the ML model in different situations. ML models performance is 50% lesser when the model is trained with synthetic and tested with physiologic ECG datasets compared to trained and tested with physiologic ECG datasets. This suggests that the ML model must be trained with physiologic ECG datasets rather than synthetic ones or add more various types of noise in synthetic ECG datasets that can mimic physiologic ECG.Clinical relevance- ML model trained with synthetic noisy ECG can increase the 50% misclassification rate in ECG noise detection compared to training with physiologic ECG datasets. The wrong classification of noise-free and noisy ECG will lead to misdiagnosis regarding the patient's condition, which could be a cause of death.
Assuntos
Inteligência Artificial , Eletrocardiografia , Humanos , Aprendizado de Máquina , Máquina de Vetores de SuporteRESUMO
The conventional approach to monitoring sleep stages requires placing multiple sensors on patients, which is inconvenient for long-term monitoring and requires expert support. We propose a single-sensor photoplethysmographic (PPG)-based automated multi-stage sleep classification. This experimental study recorded the PPG during the entire night's sleep of 10 patients. Data analysis was performed to obtain 79 features from the recordings, which were then classified according to sleep stages. The classification results using support vector machine (SVM) with the polynomial kernel yielded an overall accuracy of 84.66%, 79.62% and 72.23% for two-, three- and four-stage sleep classification. These results show that it is possible to conduct sleep stage monitoring using only PPG. These findings open the opportunities for PPG-based wearable solutions for home-based automated sleep monitoring.
RESUMO
Automatic sleep stage classification is vital for evaluating the quality of sleep. Conventionally, sleep is monitored using multiple physiological sensors that are uncomfortable for long-term monitoring and require expert intervention. In this study, we propose an automatic technique for multi-stage sleep classification using photoplethysmographic (PPG) signal. We have proposed a convolutional neural network (CNN) that learns directly from the PPG signal and classifies multiple sleep stages. We developed models for two- (Wake-Sleep), three- (Wake-NREM-REM) and four- (Wake-Light sleep-Deep sleep-REM) stages of sleep classification. Our proposed approach shows an average classification accuracy of 94.4%, 94.2%, and 92.9% for two, three, and four stages, respectively. Experimental results show that the proposed CNN model outperforms existing state-of-the-art models (classical and deep learning) in the literature.
Assuntos
Redes Neurais de Computação , Fases do Sono , Fases do Sono/fisiologia , Sono , Polissonografia , EletroencefalografiaRESUMO
The onset of stress triggers sympathetic arousal (SA), which causes detectable changes to physiological parameters such as heart rate, blood pressure, dilation of the pupils and sweat release. The objective quantification of SA has tremendous potential to prevent and manage psychological disorders. Photoplethysmography (PPG), a non-invasive method to measure skin blood flow changes, has been used to estimate SA indirectly. However, the impact of various wavelengths of the PPG signal has not been investigated for estimating SA. In this study, we explore the feasibility of using various statistical and nonlinear features derived from peak-to-peak (AC) values of PPG signals of different wavelengths (green, blue, infrared and red) to estimate stress-induced changes in SA and compare their performances. The impact of two physical stressors: and Hand Grip are studied on 32 healthy individuals. Linear (Mean, s.d.) and nonlinear (Katz, Petrosian, Higuchi, SampEn, TotalSampEn) features are extracted from the PPG signal's AC amplitudes to identify the onset, continuation and recovery phases of those stressors. The results show that the nonlinear features are the most promising in detecting stress-induced sympathetic activity. TotalSampEn feature was capable of detecting stress-induced changes in SA for all wavelengths, whereas other features (Petrosian, AvgSampEn) are significant (AUC ≥ 0.8) only for IR and Red wavelengths. The outcomes of this study can be used to make device design decisions as well as develop stress detection algorithms.