Physiological measurement最新文献

{"title":"Sex differences in chest electrical impedance tomography findings.","authors":"I Frerichs, C Händel, T Becher, D Schädler","doi":"10.1088/1361-6579/ad5ef7","DOIUrl":"10.1088/1361-6579/ad5ef7","url":null,"abstract":"<p><p><i>Objective.</i>Electrical impedance tomography (EIT) has been used to determine regional lung ventilation distribution in humans for decades, however, the effect of biological sex on the findings has hardly ever been examined. The aim of our study was to determine if the spatial distribution of ventilation assessed by EIT during quiet breathing was influenced by biological sex.<i>Approach.</i>219 adults with no known acute or chronic lung disease were examined in sitting position with the EIT electrodes placed around the lower chest (6th intercostal space). EIT data were recorded at 33 images/s during quiet breathing for 60 s. Regional tidal impedance variation was calculated in all EIT image pixels and the spatial distribution of the values was determined using the established EIT measures of centre of ventilation in ventrodorsal (CoV_vd) and right-to-left direction (CoV_rl), the dorsal and right fraction of ventilation, and ventilation defect score.<i>Main results.</i>After exclusion of one subject due to insufficient electrode contact, 218 data sets were analysed (120 men, 98 women) (age: 53 ± 18 vs 50 ± 16 yr (<i>p</i>= 0.2607), body mass index: 26.4 ± 4.0 vs 26.4 ± 6.6 kg m^-2(<i>p</i>= 0.9158), mean ± SD). Highly significant differences in ventilation distribution were identified between men and women between the right and left chest sides (CoV_rl: 47.0 ± 2.9 vs 48.8 ± 3.3% of chest diameter (<i>p</i>< 0.0001), right fraction of ventilation: 0.573 ± 0.067 vs 0.539 ± 0.071 (<i>p</i>= 0.0004)) and less significant in the ventrodorsal direction (CoV_vd: 55.6 ± 4.2 vs 54.5 ± 3.6% of chest diameter (<i>p</i>= 0.0364), dorsal fraction of ventilation: 0.650 ± 0.121 vs 0.625 ± 0.104 (<i>p</i>= 0.1155)). Ventilation defect score higher than one was found in 42.5% of men but only in 16.6% of women.<i>Significance.</i>Biological sex needs to be considered when EIT findings acquired in upright subjects in a rather caudal examination plane are interpreted. Sex differences in chest anatomy and thoracoabdominal mechanics may explain the results.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-07-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141498715","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Joint low-rank tensor fusion and cross-modal attention for multimodal physiological signals based emotion recognition.","authors":"Xin Wan, Yongxiong Wang, Zhe Wang, Yiheng Tang, Benke Liu","doi":"10.1088/1361-6579/ad5bbc","DOIUrl":"10.1088/1361-6579/ad5bbc","url":null,"abstract":"<p><p><i>Objective</i>. Physiological signals based emotion recognition is a prominent research domain in the field of human-computer interaction. Previous studies predominantly focused on unimodal data, giving limited attention to the interplay among multiple modalities. Within the scope of multimodal emotion recognition, integrating the information from diverse modalities and leveraging the complementary information are the two essential issues to obtain the robust representations.<i>Approach</i>. Thus, we propose a intermediate fusion strategy for combining low-rank tensor fusion with the cross-modal attention to enhance the fusion of electroencephalogram, electrooculogram, electromyography, and galvanic skin response. Firstly, handcrafted features from distinct modalities are individually fed to corresponding feature extractors to obtain latent features. Subsequently, low-rank tensor is fused to integrate the information by the modality interaction representation. Finally, a cross-modal attention module is employed to explore the potential relationships between the distinct latent features and modality interaction representation, and recalibrate the weights of different modalities. And the resultant representation is adopted for emotion recognition.<i>Main results</i>. Furthermore, to validate the effectiveness of the proposed method, we execute subject-independent experiments within the DEAP dataset. The proposed method has achieved the accuracies of 73.82% and 74.55% for valence and arousal classification.<i>Significance</i>. The results of extensive experiments verify the outstanding performance of the proposed method.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-07-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141451180","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhancing ECG Heartbeat classification with feature fusion neural networks and dynamic minority-biased batch weighting loss function.","authors":"Jiajun Cai, Junmei Song, Bo Peng","doi":"10.1088/1361-6579/ad5cc0","DOIUrl":"10.1088/1361-6579/ad5cc0","url":null,"abstract":"<p><p><i>Objective.</i>This study aims to address the challenges of imbalanced heartbeat classification using electrocardiogram (ECG). In this proposed novel deep-learning method, the focus is on accurately identifying minority classes in conditions characterized by significant imbalances in ECG data.<i>Approach.</i>We propose a feature fusion neural network enhanced by a dynamic minority-biased batch weighting loss function. This network comprises three specialized branches: the complete ECG data branch for a comprehensive view of ECG signals, the local QRS wave branch for detailed features of the QRS complex, and the<i>R</i>wave information branch to analyze<i>R</i>wave characteristics. This structure is designed to extract diverse aspects of ECG data. The dynamic loss function prioritizes minority classes while maintaining the recognition of majority classes, adjusting the network's learning focus without altering the original data distribution. Together, this fusion structure and adaptive loss function significantly improve the network's ability to distinguish between various heartbeat classes, enhancing the accuracy of minority class identification.<i>Main results.</i>The proposed method demonstrated balanced performance within the MIT-BIH dataset, especially for minority classes. Under the intra-patient paradigm, the accuracy, sensitivity, specificity, and positive predictive value for Supraventricular ectopic beat were 99.63%, 93.62%, 99.81%, and 92.98%, respectively, and for Fusion beat were 99.76%, 85.56%, 99.87%, and 84.16%, respectively. Under the inter-patient paradigm, these metrics were 96.56%, 89.16%, 96.84%, and 51.99%for Supraventricular ectopic beat, and 96.10%, 77.06%, 96.25%, and 13.92%for Fusion beat, respectively.<i>Significance.</i>This method effectively addresses the class imbalance in ECG datasets. By leveraging diverse ECG signal information and a novel loss function, this approach offers a promising tool for aiding in the diagnosis and treatment of cardiac conditions.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-07-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141470045","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Machine learning-based atrial fibrillation detection and onset prediction using QT-dynamicity.","authors":"Jean-Marie Grégoire, Cédric Gilon, Nathan Vaneberg, Hugues Bersini, Stéphane Carlier","doi":"10.1088/1361-6579/ad55a1","DOIUrl":"10.1088/1361-6579/ad55a1","url":null,"abstract":"<p><p><i>Objective</i>. This study examines the value of ventricular repolarization using QT dynamicity for two different types of atrial fibrillation (AF) prediction.<i>Approach</i>. We studied the importance of QT-dynamicity (1) in the detection and (2) the onset prediction (i.e. forecasting) of paroxysmal AF episodes using gradient-boosted decision trees (GBDT), an interpretable machine learning technique. We labeled 176 paroxysmal AF onsets from 88 patients in our unselected Holter recordings database containing paroxysmal AF episodes. Raw ECG signals were delineated using a wavelet-based signal processing technique. A total of 44 ECG features related to interval and wave durations and amplitude were selected and the GBDT model was trained with a Bayesian hyperparameters selection for various windows. The dataset was split into two parts at the patient level, meaning that the recordings from each patient were only present in either the train or test set, but not both. We used 80% on the database for the training and the remaining 20% for the test of the trained model. The model was evaluated using 5-fold cross-validation.<i>Main results.</i>The mean age of the patients was 75.9 ± 11.9 (range 50-99), the number of episodes per patient was 2.3 ± 2.2 (range 1-11), and CHA2DS2-VASc score was 2.9 ± 1.7 (range 1-9). For the detection of AF, we obtained an area under the receiver operating curve (AUROC) of 0.99 (CI 95% 0.98-0.99) and an accuracy of 95% using a 30 s window. Features related to RR intervals were the most influential, followed by those on QT intervals. For the AF onset forecast, we obtained an AUROC of 0.739 (0.712-0.766) and an accuracy of 74% using a 120s window. R wave amplitude and QT dynamicity as assessed by Spearman's correlation of the QT-RR slope were the best predictors.<i>Significance</i>. The QT dynamicity can be used to accurately predict the onset of AF episodes. Ventricular repolarization, as assessed by QT dynamicity, adds information that allows for better short time prediction of AF onset, compared to relying only on RR intervals and heart rate variability. Communication between the ventricles and atria is mediated by the autonomic nervous system (ANS). The variations in intraventricular conduction and ventricular repolarization changes resulting from the influence of the ANS play a role in the initiation of AF.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141288409","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Continuous monitoring methods of cerebral compliance and compensatory reserve: a scoping review of human literature.","authors":"Abrar Islam, Logan Froese, Tobias Bergmann, Alwyn Gomez, Amanjyot Singh Sainbhi, Nuray Vakitbilir, Kevin Y Stein, Izabella Marquez, Younis Ibrahim, Frederick A Zeiler","doi":"10.1088/1361-6579/ad4f4a","DOIUrl":"10.1088/1361-6579/ad4f4a","url":null,"abstract":"<p><p><i>Objective.</i>Continuous monitoring of cerebrospinal compliance (CC)<b>/</b>cerebrospinal compensatory reserve (CCR) is crucial for timely interventions and preventing more substantial deterioration in the context of acute neural injury, as it enables the early detection of abnormalities in intracranial pressure (ICP). However, to date, the literature on continuous CC/CCR monitoring is scattered and occasionally challenging to consolidate.<i>Approach.</i>We subsequently conducted a systematic scoping review of the human literature to highlight the available continuous CC/CCR monitoring methods.<i>Main results.</i>This systematic review incorporated a total number of 76 studies, covering diverse patient types and focusing on three primary continuous CC or CCR monitoring metrics and methods-Moving Pearson's correlation between ICP pulse amplitude waveform and ICP, referred to as RAP, the Spiegelberg Compliance Monitor, changes in cerebral blood flow velocity with respect to the alternation of ICP measured through transcranial doppler (TCD), changes in centroid metric, high frequency centroid (HFC) or higher harmonics centroid (HHC), and the P2/P1 ratio which are the distinct peaks of ICP pulse wave. The majority of the studies in this review encompassed RAP metric analysis (<i>n</i>= 43), followed by Spiegelberg Compliance Monitor (<i>n</i>= 11), TCD studies (<i>n</i>= 9), studies on the HFC/HHC (<i>n</i>= 5), and studies on the P2/P1 ratio studies (<i>n</i>= 6). These studies predominantly involved acute traumatic neural injury (i.e. Traumatic Brain Injury) patients and those with hydrocephalus. RAP is the most extensively studied of the five focused methods and exhibits diverse applications. However, most papers lack clarification on its clinical applicability, a circumstance that is similarly observed for the other methods.<i>Significance.</i>Future directions involve exploring RAP patterns and identifying characteristics and artifacts, investigating neuroimaging correlations with continuous CC/CCR and integrating machine learning, holding promise for simplifying CC/CCR determination. These approaches should aim to enhance the precision and accuracy of the metric, making it applicable in clinical practice.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-06-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141082050","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Identifying time-varying dynamics of heart rate and oxygen uptake from single ramp incremental running tests.","authors":"Jasper Gielen, Loes Stessens, Romain Meeusen, Jean-Marie Aerts","doi":"10.1088/1361-6579/ad56f7","DOIUrl":"10.1088/1361-6579/ad56f7","url":null,"abstract":"<p><p><i>Objective.</i>The fact that ramp incremental exercise yields quasi-linear responses for pulmonary oxygen uptake (V˙O2) and heart rate (HR) seems contradictory to the well-known non-linear behavior of underlying physiological processes. Prior research highlights this issue and demonstrates how a balancing of system gain and response time parameters causes linearV˙O2responses during ramp tests. This study builds upon this knowledge and extracts the time-varying dynamics directly from HR andV˙O2data of single ramp incremental running tests.<i>Approach.</i>A large-scale open access dataset of 735 ramp incremental running tests is analyzed. The dynamics are obtained by means of 1st order autoregressive and exogenous models with time-variant parameters. This allows for the estimates of time constant (<i>τ</i>) and steady state gain (SSG) to vary with work rate.<i>Main results.</i>As the work rate increases,<i>τ</i>-values increase on average from 38 to 132 s for HR, and from 27 to 35 s forV˙O2. Both increases are statistically significant (<i>p</i>< 0.01). Further, SSG-values decrease on average from 14 to 9 bpm (km·h^-1)^-1for HR, and from 218 to 144 ml·min^-1forV˙O2(<i>p</i>< 0.01 for decrease parameters of HR andV˙O2). The results of this modeling approach are line with literature reporting on cardiorespiratory dynamics obtained using standard procedures.<i>Significance.</i>We show that time-variant modeling is able to determine the time-varying dynamics HR andV˙O2responses to ramp incremental running directly from individual tests. The proposed method allows for gaining insights into the cardiorespiratory response characteristics when no repeated measurements are available.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":null,"pages":null},"PeriodicalIF":2.3,"publicationDate":"2024-06-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141306590","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An examination of acute physiological and perceptual responses following blood flow restriction exercise using a traditional research device or novel, automated system.","authors":"Enrique N Moreno, Elias C Figueroa, Andrew W Heath, Samuel L Buckner","doi":"10.1088/1361-6579/ad548c","DOIUrl":"10.1088/1361-6579/ad548c","url":null,"abstract":"<p><p><i>Objective</i>. To compare the acute physiological and perceptual responses to blood flow restriction (BFR) exercise using a traditional research device or novel, automated system.<i>Methods</i>. Forty-four resistance trained individuals performed four sets of unilateral elbow flexion exercise (30% one-repetition maximum) to volitional failure using two distinct restrictive devices [SmartCuffs PRO BFR Model (SMARTCUFF), Hokanson E20 Rapid Inflation device (HOKANSON)] and with two levels of BFR [40% limb occlusion pressure (LOP), 80% LOP]. Blood pressure (BP), muscle thickness (MT), and isometric strength (ISO) were assessed prior to and following exercise. Perceptual responses [ratings of perceived exertion (RPE), discomfort] were assessed prior to exercise and following each exercise set.<i>Main results</i>. Data are displayed as means (SD). Immediately following exercise with 40% LOP, there were no statistical differences between devices for BP, MT, and ISO. However, only following Set 1 of exercise, RPE was greater with SMARTCUFF compared to HOKANSON (<i>p</i>< 0.05). In addition, only following Set 2 of exercise, discomfort was greater with HOKANSON compared to SMARTCUFF (<i>p</i>< 0.001). Immediately following exercise with 80% LOP, there were no statistical differences between devices for BP, MT, and ISO. However, only following Set 4 of exercise, RPE was greater with HOKANSON compared to SMARTCUFF (<i>p</i>< 0.05). In addition, following all exercise sets, discomfort was greater with HOKANSON compared to SMARTCUFF (<i>p</i>< 0.001). For repetitions completed with 40% LOP there were no statistical differences between SMARTCUFF and HOKANSON across any exercise sets. For repetitions completed with 80% LOP there were no statistical differences between SMARTCUFF and HOKANSON across Set 1 of exercise (<i>p</i>= 0.34), however, for Sets 2-4 of exercise, significantly greater number of repetitions were completed during SMARTCUFF than HOKANSON.<i>Significance</i>. The present study provides valuable insight into the efficacy of a novel, automated BFR system (SMARTCUFF) eliciting comparable acute physiological responses to BFR exercise and in some cases favorable perceptual responses when compared to a traditional research device (HOKANSON).</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":null,"pages":null},"PeriodicalIF":3.2,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141261303","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Reconsider photoplethysmogram signal quality assessment in the free living environment.","authors":"Yan-Wei Su, Chia-Cheng Hao, Gi-Ren Liu, Yuan-Chung Sheu, Hau-Tieng Wu","doi":"10.1088/1361-6579/ad4f4b","DOIUrl":"10.1088/1361-6579/ad4f4b","url":null,"abstract":"<p><p><i>Objective.</i>Assessing signal quality is crucial for biomedical signal processing, yet a precise mathematical model for defining signal quality is often lacking, posing challenges for experts in labeling signal qualities. The situation is even worse in the free living environment.<i>Approach.</i>We propose to model a PPG signal by the adaptive non-harmonic model (ANHM) and apply a decomposition algorithm to explore its structure, based on which we advocate a reconsideration of the concept of signal quality.<i>Main results.</i>We demonstrate the necessity of this reconsideration and highlight the relationship between signal quality and signal decomposition with examples recorded from the free living environment. We also demonstrate that relying on mean and instantaneous heart rates derived from PPG signals labeled as high quality by experts without proper reconsideration might be problematic.<i>Significance.</i>A new method, distinct from visually inspecting the raw PPG signal to assess its quality, is needed. Our proposed ANHM model, combined with advanced signal processing tools, shows potential for establishing a systematic signal decomposition based signal quality assessment model.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":null,"pages":null},"PeriodicalIF":3.2,"publicationDate":"2024-06-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141082052","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Determinants of the dynamic cerebral critical closing pressure response to changes in mean arterial pressure.","authors":"Ronney B Panerai, Abdulaziz Alshehri, Lucy C Beishon, Aaron Davies, Victoria J Haunton, Emmanuel Katsogridakis, Man Y Lam, Osian Llwyd, Thompson G Robinson, Jatinder S Minhas","doi":"10.1088/1361-6579/ad548d","DOIUrl":"10.1088/1361-6579/ad548d","url":null,"abstract":"<p><p><i>Objective</i>. Cerebral critical closing pressure (CrCP) represents the value of arterial blood pressure (BP) where cerebral blood flow (CBF) becomes zero. Its dynamic response to a step change in mean BP (MAP) has been shown to reflect CBF autoregulation, but robust methods for its estimation are lacking. We aim to improve the quality of estimates of the CrCP dynamic response.<i>Approach</i>. Retrospective analysis of 437 healthy subjects (aged 18-87 years, 218 males) baseline recordings with measurements of cerebral blood velocity in the middle cerebral artery (MCAv, transcranial Doppler), non-invasive arterial BP (Finometer) and end-tidal CO₂(EtCO₂, capnography). For each cardiac cycle CrCP was estimated from the instantaneous MCAv-BP relationship. Transfer function analysis of the MAP and MCAv (MAP-MCAv) and CrCP (MAP-CrCP) allowed estimation of the corresponding step responses (SR) to changes in MAP, with the output in MCAv (SRV_MCAv) representing the autoregulation index (ARI), ranging from 0 to 9. Four main parameters were considered as potential determinants of the SRV_CrCPtemporal pattern, including the coherence function, MAP spectral power and the reconstruction error for SRV_MAP, from the other three separate SRs.<i>Main results</i>. The reconstruction error for SRV_MAPwas the main determinant of SRV_CrCPsignal quality, by removing the largest number of outliers (Grubbs test) compared to the other three parameters. SRV_CrCPshowed highly significant (<i>p</i>< 0.001) changes with time, but its amplitude or temporal pattern was not influenced by sex or age. The main physiological determinants of SRV_CrCPwere the ARI and the mean CrCP for the entire 5 min baseline period. The early phase (2-3 s) of SRV_CrCPresponse was influenced by heart rate whereas the late phase (10-14 s) was influenced by diastolic BP.<i>Significance</i>. These results should allow better planning and quality of future research and clinical trials of novel metrics of CBF regulation.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":null,"pages":null},"PeriodicalIF":3.2,"publicationDate":"2024-06-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141261729","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multimodal wearable EEG, EMG and accelerometry measurements improve the accuracy of tonic-clonic seizure detection.","authors":"Jingwei Zhang, Lauren Swinnen, Christos Chatzichristos, Victoria Broux, Renee Proost, Katrien Jansen, Benno Mahler, Nicolas Zabler, Nino Epitashvilli, Matthias Dümpelmann, Andreas Schulze-Bonhage, Elisabeth Schriewer, Ummahan Ermis, Stefan Wolking, Florian Linke, Yvonne Weber, Mkael Symmonds, Arjune Sen, Andrea Biondi, Mark P Richardson, Abuhaiba Sulaiman I, Ana Isabel Silva, Francisco Sales, Gergely Vértes, Wim Van Paesschen, Maarten De Vos","doi":"10.1088/1361-6579/ad4e94","DOIUrl":"10.1088/1361-6579/ad4e94","url":null,"abstract":"<p><p><i>Objective</i>. This paper aims to investigate the possibility of detecting tonic-clonic seizures (TCSs) with behind-the-ear, two-channel wearable electroencephalography (EEG), and to evaluate its added value to non-EEG modalities in TCS detection.<i>Methods</i>. We included 27 participants with a total of 44 TCSs from the European multicenter study SeizeIT2. The wearable Sensor Dot (Byteflies) was used to measure behind-the-ear EEG, electromyography (EMG), electrocardiography, accelerometry (ACC) and gyroscope. We evaluated automatic unimodal detection of TCSs, using sensitivity, precision, false positive rate (FPR) and F1-score. Subsequently, we fused the different modalities and again assessed performance. Algorithm-labeled segments were then provided to two experts, who annotated true positive TCSs, and discarded false positives.<i>Results</i>. Wearable EEG outperformed the other single modalities with a sensitivity of 100% and a FPR of 10.3/24 h. The combination of wearable EEG and EMG proved most clinically useful, delivering a sensitivity of 97.7%, an FPR of 0.4/24 h, a precision of 43%, and an F1-score of 59.7%. The highest overall performance was achieved through the fusion of wearable EEG, EMG, and ACC, yielding a sensitivity of 90.9%, an FPR of 0.1/24 h, a precision of 75.5%, and an F1-score of 82.5%.<i>Conclusions</i>. In TCS detection with a wearable device, combining EEG with EMG, ACC or both resulted in a remarkable reduction of FPR, while retaining a high sensitivity.<i>Significance</i>. Adding wearable EEG could further improve TCS detection, relative to extracerebral-based systems.</p>","PeriodicalId":20047,"journal":{"name":"Physiological measurement","volume":null,"pages":null},"PeriodicalIF":3.2,"publicationDate":"2024-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141076313","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}