Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 13
Encontrada(s) 1 página(s)

RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Improved Hypertension Risk Assessment with Photoplethysmographic Recordings Combining Deep Learning and Calibration

  • Cano-Serrano, Jesús
  • Bertomeu-González, Vicente
  • Fácila, Lorenzo
  • Hornenro, Fernando
  • Alcaraz, Raúl
  • Rieta, J J|||0000-0002-3364-6380
[EN] Hypertension, a primary risk factor for various cardiovascular diseases, is a global health concern. Early identification and effective management of hypertensive individuals are vital for reducing associated health risks. This study explores the potential of deep learning (DL) techniques, specifically GoogLeNet, ResNet-18, and ResNet-50, for discriminating between normotensive (NTS) and hypertensive (HTS) individuals using photoplethysmographic (PPG) recordings. The research assesses the impact of calibration at different time intervals between measurements, considering intervals less than 1 h, 1-6 h, 6-24 h, and over 24 h. Results indicate that calibration is most effective when measurements are closely spaced, with an accuracy exceeding 90% in all the DL strategies tested. For calibration intervals below 1 h, ResNet-18 achieved the highest accuracy (93.32%), sensitivity (84.09%), specificity (97.30%), and F1-score (88.36%). As the time interval between calibration and test measurements increased, classification performance gradually declined. For intervals exceeding 6 h, accuracy dropped below 81% but with all models maintaining accuracy above 71% even for intervals above 24 h. This study provides valuable insights into the feasibility of using DL for hypertension risk assessment, particularly through PPG recordings. It demonstrates that closely spaced calibration measurements can lead to highly accurate classification, emphasizing the potential for real-time applications. These findings may pave the way for advanced, non-invasive, and continuous blood pressure monitoring methods that are both efficient and reliable., This research has received financial support from public grants PID2021-123804OB-I00, PID2021-00X128525-IV0 and TED2021-130935B-I00 of the Spanish Government, jointly with the European Regional Development Fund (EU), SBPLY/17/180501/000411 and SBPLY/21/180501/000186 from Junta de Comunidades de Castilla-La Mancha, and AICO/2021/286 from Generalitat Valenciana.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A Novel Signal Restoration Method of Noisy Photoplethysmograms for Uninterrupted Health Monitoring

  • Vraka, Aikaterini
  • Zangróniz, Roberto
  • Quesada, Aurelio
  • Hornero, Fernando
  • Alcaraz, Raúl
  • Rieta, J J|||0000-0002-3364-6380
[EN] Health-tracking from photoplethysmography (PPG) signals is significantly hindered by motion artifacts (MAs). Although many algorithms exist to detect MAs, the corrupted signal often remains unexploited. This work introduces a novel method able to reconstruct noisy PPGs and facilitate uninterrupted health monitoring. The algorithm starts with spectral-based MA detection, followed by signal reconstruction by using the morphological and heart-rate variability information from the clean segments adjacent to noise. The algorithm was tested on (a) 30 noisy PPGs of a maximum 20 s noise duration and (b) 28 originally clean PPGs, after noise addition (2-120 s) (1) with and (2) without cancellation of the corresponding clean segment. Sampling frequency was 250 Hz after resampling. Noise detection was evaluated by means of accuracy, sensitivity, and specificity. For the evaluation of signal reconstruction, the heart-rate (HR) was compared via Pearson correlation (PC) and absolute error (a) between ECGs and reconstructed PPGs and (b) between original and reconstructed PPGs. Bland-Altman (BA) analysis for the differences in HR estimation on original and reconstructed segments of (b) was also performed. Noise detection accuracy was 90.91% for (a) and 99.38-100% for (b). For the PPG reconstruction, HR showed 99.31% correlation in (a) and >90% for all noise lengths in (b). Mean absolute error was 1.59 bpm for (a) and 1.26-1.82 bpm for (b). BA analysis indicated that, in most cases, 90% or more of the recordings fall within the confidence interval, regardless of the noise length. Optimal performance is achieved even for signals of noise up to 2 min, allowing for the utilization and further analysis of recordings that would otherwise be discarded. Thereby, the algorithm can be implemented in monitoring devices, assisting in uninterrupted health-tracking., This research has received financial support from public grants PID2021-123804OB-I00, PID2021-00X128525-IV0, and TED2021-130935B-I00 of the Spanish Government, 10.13039/501100011033 jointly with the European Regional Development Fund (EU), SBPLY/17/180501/000411 and SBPLY/21/180501/000186 from Junta de Comunidades de Castilla-La Mancha, and AICO/2021/286 from Generalitat Valenciana.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Novel Entropy-Based Metrics for Long-Term Atrial Fibrillation Recurrence Prediction Following Surgical Ablation: Insights from Preoperative Electrocardiographic Analysis

  • Escribano, Pilar
  • Ródenas, Juan
  • García, Manuel
  • Hornero, Fernando
  • Gracia-Baena, Juan M.
  • Alcaraz, Raúl
  • Rieta, J J|||0000-0002-3364-6380
[EN] Atrial fibrillation (AF) is a prevalent cardiac arrhythmia often treated concomitantly with other cardiac interventions through the Cox-Maze procedure. This highly invasive intervention is still linked to a long-term recurrence rate of approximately 35% in permanent AF patients. The aim of this study is to preoperatively predict long-term AF recurrence post-surgery through the analysis of atrial activity (AA) organization from non-invasive electrocardiographic (ECG) recordings. A dataset comprising ECGs from 53 patients with permanent AF who had undergone Cox-Maze concomitant surgery was analyzed. The AA was extracted from the lead V1 of these recordings and then characterized using novel predictors, such as the mean and standard deviation of the relative wavelet energy (RWEm and RWEs) across different scales, and an entropy-based metric that computes the stationary wavelet entropy variability (SWEnV). The individual predictors exhibited limited predictive capabilities to anticipate the outcome of the procedure, with the SWEnV yielding a classification accuracy (Acc) of 68.07%. However, the assessment of the RWEs for the seventh scale (RWEs7), which encompassed frequencies associated with the AA, stood out as the most promising individual predictor, with sensitivity (Se) and specificity (Sp) values of 80.83% and 67.09%, respectively, and an Acc of almost 75%. Diverse multivariate decision tree-based models were constructed for prediction, giving priority to simplicity in the interpretation of the forecasting methodology. In fact, the combination of the SWEnV and RWEs7 consistently outperformed the individual predictors and excelled in predicting post-surgery outcomes one year after the Cox-Maze procedure, with Se, Sp, and Acc values of approximately 80%, thus surpassing the results of previous studies based on anatomical predictors associated with atrial function or clinical data. These findings emphasize the crucial role of preoperative patient-specific ECG signal analysis in tailoring post-surgical care, enhancing clinical decision making, and improving long-term clinical outcomes., This research has received financial support from public grants PID2021-123804OB-I00, PID2021- 00X128525-IV0, and TED2021-130935B-I00 of the Spanish Government, 10.13039/501100011033, in conjunction with the European Regional Development Fund (EU), SBPLY/21/180501/000186, from Junta de Comunidades de Castilla-La Mancha, and AICO/2021/286 from Generalitat Valenciana. Pilar Escribano holds the 2020-PREDUCLM-15540 scholarship co-financed by the European Social Fund (ESF) operating program 2014 2020 of Castilla-La Mancha.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Arquitectura para la monitorización de la frecuencia cardiaca en tiempo real desde un robot social, Architecture for real-time heart rate monitoring from a social robot

  • Martinez-Rodrigo, Arturo
  • Castillo, José Carlos
  • Carrasco-Martínez, Sara
  • Costa, Angelo
[EN] The demographic shift towards an older population has heightened the demand for remote surveillance and assistance for seniors, particularly those living in solitude who could potentially experience stress, fear, depression and isolation. This works presents how the social robot Mini integrates a software architecture aimed at facilitating the extended independent living of elderly individuals. To accomplish this, the robot acquires data from a wearable device for real-time heart rate monitoring of senior citizens using photoplethysmography signals. These signals are decoded, parsed, processed, and classified to recognise potential heart rate anomalies and address them appropriately. When confronted with abnormally high or low pulse measurements, the robot engages with the user to cross-examine symptoms via a sequence of inquiries. If the user confirms the presence of any symptoms or fails to respond, the robot transmits an alert to caregivers by employing Telegram, a mobile device messaging service, or a dedicated visual online platform. The online platform provides monitoring interfaces designated for caregivers to deliver historical data in a comprehensive format., [ES] El cambio demográfico hacia una población de más edad ha aumentado la demanda de vigilancia y asistencia a distancia para las personas mayores, sobre todo las que viven en soledad y podrían sufrir estrés, miedo, depresión y aislamiento. Este trabajo presenta cómo el robot social Mini integra una arquitectura de software destinada a facilitar la vida independiente prolongada de las personas mayores. Para ello, el robot adquiere datos de un dispositivo wearable para la monitorización en tiempo real de la frecuencia cardiaca de personas mayores mediante señales de fotopletismografía. Estas señales se descodifican, analizan, procesan y clasifican para reconocer posibles anomalías de la frecuencia cardiaca y tratarlas adecuadamente. Ante mediciones de pulso anormalmente altas o bajas, el robot se pone en contacto con el usuario para reexaminar los síntomas mediante una secuencia de preguntas. Si el usuario confirma la presencia de algún síntoma o no responde, el robot transmite una alerta a los cuidadores empleando Telegram, un servicio de mensajería para dispositivos móviles, o una plataforma visual en línea específica. La plataforma en línea ofrece interfaces de monitorización designadas para que los cuidadores proporcionen datos históricos en un formato completo., La investigación que ha conducido a estos resultados ha recibido financiación de los proyectos: PID2021-123941OA-I00, PID2021-128525OB-I00, PID2022-140345OB-I00, TED2021- 132079B-I00 and TED2021-130935B-I00 financiado por la Agencia Estatal de Investigación (AEI), y el Ministerio de Ciencia e Innovación. Además, este trabajo ha sido financiado por la Generalitat Valenciana CIGE/ 2021/136, Regional Development and “NextGenerationEU”/PRTR, SBPLY/21/180501/000186 from Junta de Comunidades de Castilla-La Mancha.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

A Novel Nonlinear Method for Cardiovascular Data Synchronization

  • Padovano, Daniele
  • Martinez-Rodrigo, Arturo
  • Marwan, Norbert
  • Mula Muñoz, Santiago
  • Alcaraz, Raúl
  • Rieta, J J|||0000-0002-3364-6380
[EN] This study introduces and evaluates a novel nonlinear technique for synchronizing electrocardiogram (ECG) and photoplethysmogram (PPG) signals by embracing the intrinsic relationship between heart rate variability (HRV) and pulse rate variability (PRV). The proposed method utilizes normalized mutual information (NMI) derived from cross-distance matrices (CDM) of HRV and PRV, alongside the distance matrix of a reference HRV extract. We tested the methodology on two databases containing simultaneous ECG and PPG signals, including one publicly available on Physionet, and benchmarked it against various synchronization techniques, both linear (Pearson coefficient, dynamic time warping) and nonlinear (conventional mutual information, NMI with recurrence plots, cross-recurrence plot, and determinism). Results showed that the proposed method (NMI of CDM) outperformed all others, achieving synchronization rates near 50% within a 1.2-second lag threshold. This study also comprehensively examines how signal quality and recording methodology variations affect synchronization outcomes, confirming the importance of lag threshold adjustment for real accuracy assessment. However, some limitations must be kept in mind: the proposed approach is not suited for blood pressure estimation based on HRV-PRV differences, nonlinear methods generally require higher computational resources than linear ones, and further validation is needed with additional databases from real-world scenarios similar to those used in this study., This work was supported in part by MICIU/AEI/10.13039/501100011033, Fondo Europeo de Desarrollo Regional (FEDER) European Union (EU), and NextGenerationEU/Plan de Recuperacion, Transformacion y Resiliencia (PRTR) under Grant PID2021-128525OB-I00, Grant PID2021-123804OB-I00, and Grant TED2021-130935B-I00; in part by European Regional Development Fund (EU) from Junta de Comunidades de Castilla-La Mancha under Grant SBPLY/21/180501/000186 and Grant SBPLY/21/180225/000080; and in part by the Generalitat Valenciana under Grant AICO/2021/286. The work of Daniele Padovano was supported by the Predoctoral Scholarship, which is co-financed by the Operating Program of European Social Fund (ESF) (2014-2020) of Castilla-La Mancha under Grant 2022-PRED-20642.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Deep Learning and Recurrence Information Analysis for the Automatic Detection of Obstructive Sleep Apnea

  • Padovano, Daniele
  • Martinez-Rodrigo, Arturo
  • Pastor, José Manuel
  • Alcaraz, Raúl
  • Rieta, J J|||0000-0002-3364-6380
[EN] Obstructive sleep apnea (OSA) represents a significant health concern. While polysomnography (PSG) remains the gold standard, its resource-intensive nature has encouraged the exploration of further alternative approaches. Most of these were based on the heart rate variability (HRV) analysis, but only a few of them have presented a recurrence-based approach. The present paper addresses this gap by integrating convolutional neural networks (CNNs) with HRV recurrence analysis. Employing three different and publicly available databases from PhysioNet's official repository (Apnea-ECG, MIT-BIH, and UCD-DB), the presented method was able to expose concealed patterns within the distance matrix of HRV's phase space, which is discernible at an appropriate level of abstraction through CNNs. Under the challenging context of external validation (MIT-BIH and UCD for training, and Apnea-ECG for testing), the results obtained were comparable to those presented in the state of the art, achieving a peak accuracy of 75%, while maintaining balanced sensitivity and specificity at 74% and 75%, respectively. Moreover, these results obtained by the proposed CNN-based recurrence analysis of HRV also outperformed traditional time-frequency models, which have yielded values of accuracy lower than 65%. Hence, this paper highlights the importance of the proposed method in gaining new insights into OSA's HRV dynamics, offering a contribution that adds to the existing analytical approaches in the state of the art., This research has received financial support from public grants PID2021-128525OB-I00 and PID2021-123804OB-I00 of the Spanish Government 10.13039/501100011033 jointly with the European Regional Development Fund (EU), SBPLY/21/180501/000186 from Junta de Comunidades de Castilla-La Mancha, TED2021-130935B-I00, funded by the Spanish Government in conjunction with the European Regional Development Fund (EU), and AICO/2021/286 from Generalitat Valenciana. Moreover, Daniele Padovano holds a predoctoral scholarship 2022-PRED-20642, which is co-financed by the operating program of European Social Fund (ESF) 2014-2020 of Castilla-La Mancha.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Preoperative Prediction of Catheter Ablation Outcome in Persistent Atrial Fibrillation Patients through Spectral Organization Analysis of the Surface Fibrillatory Waves

  • Escribano Cano, Pilar
  • Ródenas, Juan
  • García, Manuel
  • Arias, Miguel A.
  • Hidalgo, Víctor M.
  • Calero, Sofía
  • Alcaraz, Raúl
  • Rieta, J J|||0000-0002-3364-6380
[EN] Catheter ablation (CA) is a commonly used treatment for persistent atrial fibrillation (AF). Since its medium/long-term success rate remains limited, preoperative prediction of its outcome is gaining clinical interest to optimally select candidates for the procedure. Among predictors based on the surface electrocardiogram, the dominant frequency (DF) and harmonic exponential decay (gamma) of the fibrillatory waves (f-waves) have reported promising but clinically insufficient results. Hence, the main goal of this work was to conduct a broader analysis of the f-wave harmonic spectral structure to improve CA outcome prediction through several entropy-based measures computed on different frequency bands. On a database of 151 persistent AF patients under radio-frequency CA and a follow-up of 9 months, the newly introduced parameters discriminated between patients who relapsed to AF and those who maintained SR at about 70%, which was statistically superior to the DF and approximately similar to gamma. They also provided complementary information to gamma through different combinations in multivariate models based on lineal discriminant analysis and report classification performance improvement of about 5%. These results suggest that the presence of larger harmonics and a proportionally smaller DF peak is associated with a decreased probability of AF recurrence after CA., This research received financial support from public grants PID2021-00X128525-IV0, PID2021-123804OB-I00, and TED2021-130935B-I00 of the Spanish Government 10.13039/501100011033 jointly with the European Regional Development Fund (EU), SBPLY/17/180501/000411 and SBPLY/21/180501/000186 from Junta de Comunidades de Castilla-La Mancha, and AICO/2021/286 from Generalitat Valenciana. Moreover, Pilar Escribano holds a predoctoral scholarship 2020PREDUCLM-15540, which is co-financed by the operating program of the European Social Fund (ESF) 2014-2020 of Castilla-La Mancha.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Combination of frequency- and time-domain characteristics of the fibrillatory waves for enhanced prediction of persistent atrial fibrillation recurrence after catheter ablation

  • Escribano, Pilar
  • Ródenas, Juan
  • García, Manuel
  • Arias, Miguel A.
  • Hidalgo, Víctor M.
  • Calero, Sofía
  • Alcaraz, Raúl
  • Rieta, J J|||0000-0002-3364-6380
[EN] Catheter ablation (CA) remains the cornerstone alternative to cardioversion for sinus rhythm (SR) restoration in patients with atrial fibrillation (AF). Unfortunately, despite the last methodological and technological advances, this procedure is not consistently effective in treating persistent AF. Beyond introducing new indices to characterize the fibrillatory waves (f -waves) recorded through the preoperative electrocardiogram (ECG), the aim of this study is to combine frequency- and time -domain features to improve CA outcome prediction and optimize patient selection for the procedure, given the absence of any study that jointly analyzes information from both domains. Precisely, the f -waves of 151 persistent AF patients undergoing their first CA procedure were extracted from standard V1 lead. Novel spectral and amplitude features were derived from these waves and combined through a machine learning algorithm to anticipate the intervention midterm outcome. The power rate index (phi), which estimates the power of the harmonic content regarding the dominant frequency (DF), yielded the maximum individual discriminant ability of 64% to discern between individuals who experienced a recurrence of AF and those who sustained SR after a 9 -month follow-up period. The predictive accuracy was improved up to 78.5% when this parameter phi was merged with the amplitude spectrum area in the DF bandwidth (AMSALF) and the normalized amplitude of the f -waves into a prediction model based on an ensemble classifier, built by random undersampling boosting of decision trees. This outcome suggests that the synthesis of both spectral and temporal features of the f -waves before CA might enrich the prognostic knowledge of this therapy for persistent AF patients., This research was financially supported from public grants PID2021-00X128525-IV0, PID2021-123804OB-I00, and TED2021-130935B-I00 of the Spanish Government 10.13 039/501100011033 jointly with the European Regional Development Fund (EU) , SBPLY/21/180501/000186 from Junta de Comunidades de Castilla-La Mancha, and AICO/2021/286 from Generalitat Valenciana. Furthermore, Pilar Escribano holds a 2020-PREDUCLM-15540 predoctoral scholarship, co-financed by the operating program of the European Social Fund (ESF) 2014-2020 of Castilla-La Mancha.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Electrical Cardioversion Outcome Prognosis: A Multivariate Multiscale Entropy Characterization of Atrial Activity in Persistent Atrial Fibrillation

  • Cirugeda Roldan, Eva Maria
  • Plancha, Eva
  • Hidalgo, Victor M.
  • Calero, Sofía
  • Alcaraz, Raúl
  • Rieta, J J|||0000-0002-3364-6380
[EN] Background: Atrial fibrillation (AF) remains a significant cause of stroke, heart failure, and cardiovascular morbidity worldwide. Despite advancements in AF management, electrical cardioversion (ECV) remains the most commonly used technique for sinus rhythm (SR) restoration although presenting a limited success rate in the mid-term along with a high number of side-effects which can lead to an increase in patients health deterioration and, consequently, in healthcare costs. Hence, predicting ECV outcome in the mid-term remains a challenging task. Here, a new framework based on multivariate multiscale entropy (MMSE) characterization of atrial activity is proposed to improve ECV outcome prediction in the mid-term. Methods: 58 patients with persistent AF scheduled for ECV were considered. A 12-Lead standard ECG segment of 1.5 min duration prior to the first electrical shock was analyzed. The atrial activity (AA) is estimated from the 12-lead surface ECG using a QT segment removal algorithm based on QRS complex estimation and pattern recognition techniques. AA is characterized by means of multivariate extensions of traditional indices such as the amplitude of the fibrillatory waves and dominant frequency along with multivariate extensions of complexity measures as multivariate Sample Entropy and finally Multivariate Multiscale Entropy (MMSE). These indices were estimated over 12-lead ECG records from 58 ECV derived patients who were classified based on SR maintenance after 30-day follow up (mid-term evaluation). ECV prognosis was evaluated using ROC curves and Youden's Criteria for optimal threshold establishment. Performance was compared to that of unidimensional indices. Results: Patients who maintained SR post-ECV exhibited distinct complexity patterns compared to those who relapsed into AF. Specifically, MMSE provided higher discriminant accuracy than traditional unidimensional indices. When considering only the limb leads in the analysis, the best performance was achieved, over 83% accurate classification of SR restoration in the mid-term (Se = 0.74, Sp = 0.85, p <= 0.001). Additionally, the accumulated entropy and slope of the MMSE curves, offered robust predictors of ECV outcomes providing better balanced sensitivity and specificity ROC curves. Conclusions: This work highlights the importance of multivariate approaches in AF characterization and provides a comprehensive framework for improving ECV outcome prediction, providing an increase in almost a 30% of correct predictions in the mid-term. Future research should explore the integration of these methods into clinical practice to optimize treatment strategies for AF patients and reduced healthcare costs. (c) 2025 AGBM. Published by Elsevier Masson SAS. This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/)., This work has been supported by: projects: 1) PID2021-128525OB-I00, PID2021-123804OB-I00 and TED2021-130935B-I00 of the Spanish Government 10.13039/501100011033 jointly with the European Regional Development Fund 2) SBPLY/21/180501/000186 from Junta de Comunidades de Castilla-La Mancha 3) CIAICO/2023/200 from Generalitat Valenciana.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Single-lead electrocardiogram quality assessment in the context of paroxysmal atrial fibrillation through phase space plots

  • Huerta, Alvaro
  • Martínez-Rodrigo, Arturo
  • Bertomeu-González, Vicente
  • Ayo-Martin, Oscar
  • Alcaraz, Raúl
  • Rieta, J J|||0000-0002-3364-6380
[EN] Current wearable electrocardiogram (ECG) recording systems have great potential to revolutionize early diagnosis of paroxysmal atrial fibrillation (AF). They are able to continuously acquire an ECG signal for long weeks and then increase the probability of detecting first brief, intermittent signs of the arrhythmia. However, the recorded signal is often broadly corrupted by noise and artifacts, and accurate assessment of its quality to avoid automated misdiagnosis and false alarms of AF is still an unsolved challenge. In this context, the present work is pioneer in exploring the usefulness of transforming the single-lead ECG signal into two common phase space (PS) representations, such as the Poincare plot and the first order difference graph, for evaluation of its quality. Several machine and deep learning models fed with features and images derived from these PS portraits reported a better performance than well-known previous methods, even when they were trained and validated on two separate databases. Indeed, in binary classification of high- and low-quality ECG excerpts, the generated PS-based algorithms reported a discriminant power greater than 85%, misclassifying less than 20% of high-quality AF episodes and non -normal rhythms as noisy excerpts. Moreover, because both PS reconstructions do not require any mathematical transformation, these algorithms also spent much less time in classifying each ECG excerpt in validation and testing stages than previous methods. As a consequence, ECG transformation to both PS portraits enables novel, simple, effective, and computational low-cost techniques, based both on machine and deep learning classifiers, for ECG quality assessment., This research has received financial support from Daiichi Sankyo SLU and from public grants PID2021-00X128525-IV0, PID2021-12380 4OB-I00, and TED2021-130935B-I00 of the Spanish Government 10.13039/501100011033 jointly with the European Regional Development Fund (EU) , SBPLY/21/180501/000186 from Junta de Comunidades de Castilla-La Mancha, Spain, and AICO/2021/286 from Generalitat Valenciana.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Comparative Study on the Generalization Ability of Machine Learning and Deep Learning Algorithms for Quality Assessment of Wearable PPG Recordings

  • Mula Muñoz, Santiago
  • Zangróniz, Roberto
  • Ayo-Martin, Oscar
  • Alcaraz, Raúl
  • Rieta, J J|||0000-0002-3364-6380
[EN] One of the major challenges in using photoplethysmography (PPG) sensors for heart rate monitoring in real-world settings is ensuring signal quality. This work evaluates and compares quality assessment methods using generic machine learning (ML) and deep learning (DL) pipelines, on a unique and comprehensive framework that includes different sensors, wavelengths, measurement locations, and recording environments. The PPG signals from one proprietary and five publicly available datasets were labeled in terms of quality by comparing the PPG-derived heart rate to a reference heart rate estimated from simultaneous electrocardiograms. Diverse techniques based on common ML classifiers and one- and two-dimensional convolutional neural networks (CNN) were trained on a dataset and tested on the remaining ones. The results showed that several generated models performed comparably to previous studies when they were tested on datasets with similar measurement positions and sensors to the training database. Specifically, reductions in sensitivity, specificity, and F1-score of less than 3% from training to testing were observed on some methods. Contrarily, they reported a notably poorer performance when tested on datasets presenting conditions different from the training. Even the best-performing model, based on the well-known, pre-trained CNN AlexNet, experienced a performance drop of over 20% in that situation. These findings show that the analyzed ML and DL methods lack the ability to generalize across PPG signals captured from diverse environments, sensors, wavelengths, and measurement locations. This suggests that developing case-specific methods might be the shortest path towards reliable PPG quality assessment., This work has received financial support from Daiichi Sankyo Company, Limited and from public grants PID2021-128525OB-I00, PID2021-123804OB-I00, and TED2021-130935B-I00 of the Spanish Government 10.13039/501100011033 jointly with the European Regional Development Fund (EU), SB-PLY/21/180501/000186 from Junta de Comunidades de Castilla-La Mancha, and CIAICO/2023/200 from Generalitat Valenciana.
RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Publicación Artículo científico (JournalArticle)

Sinus rhythm maintenance in persistent atrial fibrillation: 12-lead ECG multiscale entropy characterization

  • Cirugeda, Eva M.
  • Plancha, Eva
  • Hidalgo, Victor M.
  • Calero, Sofia
  • Alcaraz, Raúl
  • Rieta, J J|||0000-0002-3364-6380
[EN] Persistent atrial fibrillation is the most common sustained cardiac arrhythmia, frequently linked with increased mortality and morbidity. Electrical cardioversion (ECV) remains the gold standard for sinus rhythm (SR) restoration, even though presenting potential adverse effects and a high relapsing rate. Predicting ECV outcome from the 12-lead ECG could reduce healthcare costs while preventing complications in patients unlikely to maintain SR. To this end, atrial activity (AA) organization has been traditionally evaluated through the amplitude and dominant frequency of the fibrillatory waves at lead II. However, physiological systems are known to exhibit complex dynamics across multiple time-scales, making multiscale (MSE) entropy measures a more suitable tool, as they can incorporate relevant information that may have been previously overlooked. Here, the predictive power of different MSE-based indices for the ECV outcome in 58 patients is evaluated. AA was estimated using a QT segment cancellation algorithm. Patients were classified based on SR maintenance after a 30-day follow-up. Results show that traditionally used indices report the highest predictive rate over the limb leads (79%). However, they are outperformed by Refined MSE over precordial leads (87%). Moreover, when considering statistical modeling techniques such as support vector machines, the prediction accuracy is increased (98%). In conclusion, MSE-based indices computed from precordial leads can robustly predict ECV outcome with higher accuracy than traditional approaches., This research has received financial support from public grants PID2021-128525OB-I00, PID2021-123804OB-I00, and TED2021-130935B-I00 of the Spanish Government 10.13039/501100011033 jointly with the European Regional Development Fund (EU), SBPLY/21/180501/000186 from Junta de Comunidades de Castilla-La Mancha, and CIAICO/2023/200 from Generalitat Valenciana.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Arquitectura para la monitorización de la frecuencia cardíaca en tiempo real desde un robot social, Architecture for real-time heart rate monitoring from a social robot

  • Martinez-Rodrigo, Arturo
  • Castillo, José Carlos
  • Carrasco-Martínez, Sara
  • Costa, Angelo
  • Costa, Angelo|||/items/f4ecad24-bcc1-47f6-a263-6b525558ae6c
El cambio demográfico hacia una población de mayor edad ha aumentado la demanda de vigilancia y asistencia a distancia para las personas mayores, especialmente aquellas que viven en soledad y podrían sufrir estrés, miedo, depresión y aislamiento. Este trabajo presenta cómo el robot social Mini integra una arquitectura de software destinada a facilitar la vida independiente prolongada de las personas mayores. Para ello, el robot adquiere datos de un dispositivo wearable para la monitorización en tiempo real de la frecuencia cardíaca de personas mayores mediante señales de fotopletismografía. Estas señales se descodifican, analizan, procesan y clasifican para reconocer posibles anomalías en la frecuencia cardíaca y tratarlas adecuadamente. Ante mediciones de pulso anormalmente altas o bajas, el robot se pone en contacto con el usuario para reexaminar los síntomas mediante una secuencia de preguntas. Si el usuario confirma la presencia de algún síntoma o no responde, el robot transmite una alerta a los cuidadores empleando Telegram, un servicio de mensajería para dispositivos móviles, o una plataforma visual en línea específica. La plataforma en línea ofrece interfaces de monitorización designadas para que los cuidadores proporcionen datos históricos en un formato completo.