Sensors

Research

18 pages, 8725 KB

Open AccessArticle

Assessment of Anesthetic Depth Through EEG Mode Decomposition Using Singular Spectrum Analysis

by Haruka Kida, Tomomi Yamada, Shoko Yamochi, Yurie Obata, Fumimasa Amaya and Teiji Sawa

Sensors 2026, 26(4), 1212; https://doi.org/10.3390/s26041212 - 12 Feb 2026

Viewed by 380

(1) Background: Electroencephalography (EEG) is widely used to monitor the depth of anesthesia; however, conventional Fourier-based analyses are limited in their ability to characterize non-stationary anesthetic-induced EEG dynamics. In this study, we investigated the utility of singular spectrum analysis (SSA) combined with the [...] Read more.

(1) Background: Electroencephalography (EEG) is widely used to monitor the depth of anesthesia; however, conventional Fourier-based analyses are limited in their ability to characterize non-stationary anesthetic-induced EEG dynamics. In this study, we investigated the utility of singular spectrum analysis (SSA) combined with the Hilbert transform for extracting physiologically meaningful EEG features under sevoflurane general anesthesia. (2) Methods: Frontal EEG data from ten patients undergoing sevoflurane anesthesia were analyzed from the maintenance phase through emergence. Using SSA, short EEG segments were decomposed into six intrinsic mode functions (IMFs) without pre-specified basis functions or frequency bands. Hilbert spectral analysis was applied to each IMF to obtain instantaneous frequency and amplitude characteristics. (3) Results: The SSA-based decomposition clearly captured phase-dependent EEG changes, including α spindle activity during maintenance and increasing high-frequency components preceding emergence. Multiple linear regression models incorporating IMF center frequencies and total power demonstrated strong correlations with the bispectral index (BIS), achieving high predictive accuracy (R² = 0.88, MAE < 4). Compared with conventional spectral approaches, SSA provided superior temporal resolution and stable feature extraction for non-stationary EEG signals. (4) Conclusions: These findings indicate that SSA combined with Hilbert analysis is a robust framework for quantitative EEG analysis during general anesthesia and may enhance real-time, individualized assessments of anesthetic depth. Full article

(This article belongs to the Special Issue Advances in ECG/EEG Monitoring)

► Show Figures

Graphical abstract

28 pages, 5694 KB

Open AccessArticle

How the Level of Noise Affects Temporal Accuracy of a QRS Detector—Case Study

by Wojciech Reklewski and Piotr Augustyniak

Sensors 2026, 26(1), 15; https://doi.org/10.3390/s26010015 - 19 Dec 2025

Viewed by 393

Abstract

Background: QRS complex detection is a key processing step of automated ECG analysis and determines its overall quality. The purpose of this paper is to study the detection performance of probably the most frequently implemented ready-to-use QRS detector in the presence of noise [...] Read more.

Background: QRS complex detection is a key processing step of automated ECG analysis and determines its overall quality. The purpose of this paper is to study the detection performance of probably the most frequently implemented ready-to-use QRS detector in the presence of noise and with tightened temporal tolerance of detection points. Methods: We applied commonly used detection statistics (Detection Error Rate, Sensitivity, Positive Predictive Value, and F₁ score), but re-defined true positive detection based on variable time jitter between detected and reference points. We also applied a controlled level of mixed noise to assess the detector’s performance in true-to-life conditions. Results: We found the following: (1) the detector under test showed a considerable drop in quality when reducing the jitter between 97.23 ms (DER = 8.08%) and 86.12 ms (DER = 67.22%), which means that the detection points’ time series are not accurate enough to be directly used for ECG time analysis; (2) with jitter allowed to 163.90 ms and an increasing noise level (SNR from 20 dB to −7.96 dB), the detection quality drops (DER from 0.98% to 57.13% respectively); however, an analysis of individual files revealed records, where the algorithm performs better in the presence of noise; (3) with a step-by-step code execution analysis of ECG strips where better performance was the most prominent, the imprecise definition of the local maximum was the cause of DER errors. Conclusions: Our research clearly indicates that selecting a QRS-detection algorithm based solely on DER, Se, and PPV detection statistics may be incorrect. Two equally important detection quality parameters are the change in the DER error rate with tightened requirements of jitter and robustness of the detection statistics DER, Se, and PPV to noise level variations (algorithm’s detection points immunity to noise). Full article

(This article belongs to the Special Issue Advances in ECG/EEG Monitoring)

► Show Figures

Figure 1

27 pages, 4157 KB

Open AccessArticle

ECG-Based Detection of Epileptic Seizures in Real-World Wearable Settings: Insights from the SeizeIT2 Dataset

by Conrad Reintjes, Janosch Fabio Hagenbeck, Mohamed Ballo, Tim Rahlmeier, Simon Maximilian Wolf and Detlef Schoder

Sensors 2025, 25(24), 7687; https://doi.org/10.3390/s25247687 - 18 Dec 2025

Cited by 1 | Viewed by 1308

Abstract

Epilepsy is a prevalent neurological disorder where reliable seizure tracking is essential for patient care. Existing documentation often relies on self-reports, which are unreliable, creating a need for objective, wearable-based solutions. Prior work has shown that Electrocardiography (ECG)-based seizure detection is feasible but [...] Read more.

Epilepsy is a prevalent neurological disorder where reliable seizure tracking is essential for patient care. Existing documentation often relies on self-reports, which are unreliable, creating a need for objective, wearable-based solutions. Prior work has shown that Electrocardiography (ECG)-based seizure detection is feasible but limited by small datasets. This study addresses this issue by evaluating Matrix Profile, MADRID, and TimeVQVAE-AD on SeizeIT2, the largest open wearable-ECG dataset with 11,640 recording hours and 886 annotated seizures. Using standardized preprocessing and clinically motivated windows, we benchmarked sensitivity, false-alarm rate (FAR), and a Harmonic Mean Score integrating both metrics. Across methods, TimeVQVAE-AD achieved the highest sensitivity, while MADRID produced the lowest FAR, illustrating the trade-off between detecting seizures and minimizing spurious alerts. Our findings show ECG anomaly detection on SeizeIT2 can reach clinically meaningful sensitivity while highlighting the sensitivity–false alarm trade-off. By releasing reproducible benchmarks and code, this work establishes the first open baseline and enables future research on personalization and clinical applicability. Full article

(This article belongs to the Special Issue Advances in ECG/EEG Monitoring)

► Show Figures

Figure 1

17 pages, 4091 KB

Open AccessArticle

EEG-Based Prediction of Stress Responses to Naturalistic Decision-Making Stimuli in Police Cadets

by Abdulwahab Alasfour and Nasser AlSabah

Sensors 2025, 25(18), 5925; https://doi.org/10.3390/s25185925 - 22 Sep 2025

Cited by 1 | Viewed by 1572

Abstract

The ability of police officers to make correct decisions under emotional stress is critical, as errors in high-pressure situations can have severe legal and physical consequences. This study aims to evaluate the neurophysiological responses of police academy cadets during stressful decision-making scenarios and [...] Read more.

The ability of police officers to make correct decisions under emotional stress is critical, as errors in high-pressure situations can have severe legal and physical consequences. This study aims to evaluate the neurophysiological responses of police academy cadets during stressful decision-making scenarios and to predict individual stress levels from those responses. Fifty-eight police academy cadets from three cohorts watched a custom-made, naturalistic video scene and then chose the appropriate course of action. Simultaneous 32-channel electroencephalography (EEG) and electrocardiography (ECG) captured brain and heart activity. Event-related potentials (ERPs) and band-specific power features (particularly delta) were extracted, and machine-learning models were trained with nested cross-validation to predict perceived stress scores. Global and broadband EEG activity was suppressed during the video stimulus and did not return to baseline during the cooldown phase. Widespread ERPs and pronounced delta-band dynamics emerged during decision-making, correlating with both cohort rank and self-reported stress. Crucially, a combined EEG + cohort model predicted perceived stress with an out-of-fold R² of 0.32, outperforming EEG-only (R² = 0.23) and cohort-only (R² = 0.17) models. To our knowledge, this is the first study to both characterize EEG dynamics during stressful naturalistic decision tasks and demonstrate their predictive utility. These findings lay the groundwork for neurofeedback-based training paradigms that help officers modulate stress responses and calibrate decision-making under pressure. Full article

(This article belongs to the Special Issue Advances in ECG/EEG Monitoring)

► Show Figures

Figure 1

23 pages, 1705 KB

Open AccessArticle

Personalizing Seizure Detection for Individual Patients by Optimal Selection of EEG Signals

by Rosanna Ferrara, Martino Giaquinto, Gennaro Percannella, Leonardo Rundo and Alessia Saggese

Sensors 2025, 25(9), 2715; https://doi.org/10.3390/s25092715 - 25 Apr 2025

Cited by 4 | Viewed by 2075

Abstract

Electroencephalography is a widely used non-invasive method for monitoring brain electrical activity, critical for diagnosing and managing neurological disorders such as epilepsy. While clinical standards use 21 electrodes to capture comprehensive neural signals, a personalized approach can enhance performance by selecting patient-specific channels, [...] Read more.

Electroencephalography is a widely used non-invasive method for monitoring brain electrical activity, critical for diagnosing and managing neurological disorders such as epilepsy. While clinical standards use 21 electrodes to capture comprehensive neural signals, a personalized approach can enhance performance by selecting patient-specific channels, reducing noise and redundancy. This study introduces an innovative, lightweight deep learning system optimized for real-time seizure detection in personalized wearable devices. The system uses an efficient Convolutional Neural Network that processes data from just two channels. These channels are automatically selected using a data-driven mechanism that identifies the most informative scalp regions based on each patient’s unique seizure patterns. The proposed approach ensures high reliability, even with small datasets, and improves interpretability for clinicians by overcoming the limitations of more complex methods. The tailored channel selection boosts detection accuracy and ensures robust performance across different seizure types while reducing the computational burden typical of multi-electrode systems. Validation on the publicly available CHB-MIT dataset achieved an average balanced accuracy of 0.83 and a false-positive rate of approximately 0.1/h. The system’s performance matches, and in some cases outperforms, state-of-the-art systems that use four fixed channels in temporal regions, demonstrating the potential of two-channel wearable solutions, specifically with a non-negligible 30% reduction in the false-positive rate. This interpretable, patient-specific method enables the development of personalized, efficient, and compact wearable devices for reliable seizure detection in everyday life. Full article

(This article belongs to the Special Issue Advances in ECG/EEG Monitoring)

► Show Figures

Figure 1

18 pages, 3238 KB

Open AccessArticle

Reversible Watermarking for Electrocardiogram Protection

by Pavel Andreev, Anna Denisova and Victor Fedoseev

Sensors 2025, 25(7), 2185; https://doi.org/10.3390/s25072185 - 30 Mar 2025

Cited by 1 | Viewed by 891

Abstract

The electrocardiogram (ECG) is one of the widespread diagnostic methods used in telemedicine. However, in the telemedicine systems, the data transfer process to the end user may suffer from security risks. Reversible watermarking can preserve the security of electrocardiograms and keep their original [...] Read more.

The electrocardiogram (ECG) is one of the widespread diagnostic methods used in telemedicine. However, in the telemedicine systems, the data transfer process to the end user may suffer from security risks. Reversible watermarking can preserve the security of electrocardiograms and keep their original precision for correct diagnostics. In this paper, we present an extensive investigation of four reversible watermarking methods: prediction error expansion (PEE), reversible contrast mapping difference expansion (RCM), integer transform-based difference expansion (ITB), and compression-based watermarking. We discover new facets of the existing ECG watermarking methods (PEE and compression-based watermarking) and adapt image watermarking methods (RCM and ITB) to ECG signal. We compare different kinds of prediction and compression methods used in the studied methods and provide a watermark capacity comparison for different methods’ implementations. The research results will help in watermarking method selection in practice. Full article

(This article belongs to the Special Issue Advances in ECG/EEG Monitoring)

► Show Figures

Figure 1

19 pages, 13145 KB

Open AccessArticle

AI-Powered Noninvasive Electrocardiographic Imaging Using the Priori-to-Attention Network (P2AN) for Wearable Health Monitoring

by Shijie He, Hanrui Dong, Xianbin Zhang, Richard Millham, Lin Xu and Wanqing Wu

Sensors 2025, 25(6), 1810; https://doi.org/10.3390/s25061810 - 14 Mar 2025

Viewed by 2021

Abstract

The rapid development of smart wearable devices has significantly advanced noninvasive, continuous health monitoring, enabling real-time collection of vital biosignals. Electrocardiographic imaging (ECGI), a noninvasive technique that reconstructs transmembrane potential (TMP) from body surface potential, has emerged as a promising method for reflecting [...] Read more.

The rapid development of smart wearable devices has significantly advanced noninvasive, continuous health monitoring, enabling real-time collection of vital biosignals. Electrocardiographic imaging (ECGI), a noninvasive technique that reconstructs transmembrane potential (TMP) from body surface potential, has emerged as a promising method for reflecting cardiac electrical activity. However, the ECG inverse problem’s inherent instability has hindered its practical application. To address this, we introduce a novel Priori-to-Attention Network (P2AN) that enhances the stability of ECGI solutions. By leveraging the one-dimensional nature of electrical signals and the body’s electrical propagation properties, P2AN uses small-scale convolutions for attention computation, integrating a priori physiological knowledge via cross-attention mechanisms. This approach eliminates the need for clinical TMP measurements and improves solution accuracy through normalization constraints. We evaluate the method’s effectiveness in diagnosing myocardial ischemia and ventricular hypertrophy, demonstrating significant improvements in TMP reconstruction and lesion localization. Moreover, P2AN exhibits high robustness in noisy environments, making it highly suitable for integration with wearable electrocardiographic clothing. By improving spatiotemporal accuracy and noise resilience, P2AN offers a promising solution for noninvasive, real-time cardiovascular monitoring using AI-powered wearable devices. Full article

(This article belongs to the Special Issue Advances in ECG/EEG Monitoring)

► Show Figures

Figure 1

22 pages, 873 KB

Open AccessArticle

EEG-Based Music Emotion Prediction Using Supervised Feature Extraction for MIDI Generation

by Oscar Gomez-Morales, Hernan Perez-Nastar, Andrés Marino Álvarez-Meza, Héctor Torres-Cardona and Germán Castellanos-Dominguez

Sensors 2025, 25(5), 1471; https://doi.org/10.3390/s25051471 - 27 Feb 2025

Cited by 5 | Viewed by 4053

Abstract

Advancements in music emotion prediction are driving AI-driven algorithmic composition, enabling the generation of complex melodies. However, bridging neural and auditory domains remains challenging due to the semantic gap between brain-derived low-level features and high-level musical concepts, making alignment computationally demanding. This study [...] Read more.

Advancements in music emotion prediction are driving AI-driven algorithmic composition, enabling the generation of complex melodies. However, bridging neural and auditory domains remains challenging due to the semantic gap between brain-derived low-level features and high-level musical concepts, making alignment computationally demanding. This study proposes a deep learning framework for generating MIDI sequences aligned with labeled emotion predictions through supervised feature extraction from neural and auditory domains. EEGNet is employed to process neural data, while an autoencoder-based piano algorithm handles auditory data. To address modality heterogeneity, Centered Kernel Alignment is incorporated to enhance the separation of emotional states. Furthermore, regression between feature domains is applied to reduce intra-subject variability in extracted Electroencephalography (EEG) patterns, followed by the clustering of latent auditory representations into denser partitions to improve MIDI reconstruction quality. Using musical metrics, evaluation on real-world data shows that the proposed approach improves emotion classification (namely, between arousal and valence) and the system’s ability to produce MIDI sequences that better preserve temporal alignment, tonal consistency, and structural integrity. Subject-specific analysis reveals that subjects with stronger imagery paradigms produced higher-quality MIDI outputs, as their neural patterns aligned more closely with the training data. In contrast, subjects with weaker performance exhibited auditory data that were less consistent. Full article

(This article belongs to the Special Issue Advances in ECG/EEG Monitoring)

► Show Figures

Figure 1

28 pages, 4312 KB

Open AccessArticle

Application of Convolutional Neural Network for Decoding of 12-Lead Electrocardiogram from a Frequency-Modulated Audio Stream (Sonified ECG)

by Vessela Krasteva, Ivo Iliev and Serafim Tabakov

Sensors 2024, 24(6), 1883; https://doi.org/10.3390/s24061883 - 15 Mar 2024

Cited by 5 | Viewed by 3949

Abstract

Research of novel biosignal modalities with application to remote patient monitoring is a subject of state-of-the-art developments. This study is focused on sonified ECG modality, which can be transmitted as an acoustic wave and received by GSM (Global System for Mobile Communications) microphones. [...] Read more.

Research of novel biosignal modalities with application to remote patient monitoring is a subject of state-of-the-art developments. This study is focused on sonified ECG modality, which can be transmitted as an acoustic wave and received by GSM (Global System for Mobile Communications) microphones. Thus, the wireless connection between the patient module and the cloud server can be provided over an audio channel, such as a standard telephone call or audio message. Patients, especially the elderly or visually impaired, can benefit from ECG sonification because the wireless interface is readily available, facilitating the communication and transmission of secure ECG data from the patient monitoring device to the remote server. The aim of this study is to develop an AI-driven algorithm for 12-lead ECG sonification to support diagnostic reliability in the signal processing chain of the audio ECG stream. Our methods present the design of two algorithms: (1) a transformer (ECG-to-Audio) based on the frequency modulation (FM) of eight independent ECG leads in the very low frequency band (300–2700 Hz); and (2) a transformer (Audio-to-ECG) based on a four-layer 1D convolutional neural network (CNN) to decode the audio ECG stream (10 s @ 11 kHz) to the original eight-lead ECG (10 s @ 250 Hz). The CNN model is trained in unsupervised regression mode, searching for the minimum error between the transformed and original ECG signals. The results are reported using the PTB-XL 12-lead ECG database (21,837 recordings), split 50:50 for training and test. The quality of FM-modulated ECG audio is monitored by short-time Fourier transform, and examples are illustrated in this paper and supplementary audio files. The errors of the reconstructed ECG are estimated by a popular ECG diagnostic toolbox. They are substantially low in all ECG leads: amplitude error (quartile range RMSE = 3–7 μV, PRD = 2–5.2%), QRS detector (Se, PPV > 99.7%), P-QRS-T fiducial points’ time deviation (<2 ms). Low errors generalized across diverse patients and arrhythmias are a testament to the efficacy of the developments. They support 12-lead ECG sonification as a wireless interface to provide reliable data for diagnostic measurements by automated tools or medical experts. Full article

(This article belongs to the Special Issue Advances in ECG/EEG Monitoring)

► Show Figures

Figure 1

13 pages, 5333 KB

Open AccessCommunication

A High-Performance System for Weak ECG Real-Time Detection

by Kun Xu, Yi Yang, Yu Li, Yahui Zhang and Limin Zhang

Sensors 2024, 24(4), 1088; https://doi.org/10.3390/s24041088 - 7 Feb 2024

Cited by 2 | Viewed by 4019

Abstract

Wearable devices have been widely used for the home monitoring of physical activities and healthcare conditions, among which ambulatory electrocardiogram (ECG) stands out for the diagnostic cardiovascular information it contains. Continuous and unobtrusive sensing often requires the integration of wearable sensors to existing [...] Read more.

Wearable devices have been widely used for the home monitoring of physical activities and healthcare conditions, among which ambulatory electrocardiogram (ECG) stands out for the diagnostic cardiovascular information it contains. Continuous and unobtrusive sensing often requires the integration of wearable sensors to existing devices such as watches, armband, headphones, etc.; nonetheless, it is difficult to detect high-quality ECG due to the nature of low signal amplitude at these areas. In this paper, a high-performance system with multi-channel signal superposition for weak ECG real-time detection is proposed. Firstly, theoretical analysis and simulation is performed to demonstrate the effectiveness of this system design. The detection system, including electrode array, acquisition board, and the application (APP), is then developed and the electrical characteristics are measured. A common mode rejection ratio (CMRR) of up to 100 dB and input inferred voltage noise below 1 μV are realized. Finally, the technique is implemented in form of ear-worn and armband devices, achieving an SNR over 20 dB. Results are also compared with the simultaneous recording of standard lead I ECG. The correlation between the heart rates derived from experimental and standard signals is higher than 0.99, showing the feasibility of the proposed technique. Full article

(This article belongs to the Special Issue Advances in ECG/EEG Monitoring)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Advances in ECG/EEG Monitoring

Share This Special Issue

Special Issue Editor

Special Issue Information

Keywords

Benefits of Publishing in a Special Issue

Published Papers (10 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI