María Fernanda Rodríguez, A. Ravelo-García, E. Alvarez, Luz Alexandra Díaz, D. Cornejo, Victor Cabrera-Caso, Dante Condori-Merma, Miguel Vizcardo Cornejo
It is estimated that in the world there are between 6 and 8 million people infected with Chagas disease, mainly in endemic areas of 21 Latin American countries, and in recent years it is slowly becoming a health problem in more urban areas and countries. In that sense, developing diagnosis methods is primordial. That is why this work used a deep neural network to classify 292 subjects (volunteers and patients) composed of 83 health volunteers (Control group); 102 asymptomatic chagasic patients (CH1 group) and 107 seropositive chagasic patients with incipient heart disease (CH2 group). Approximate Entropy ApEn was calculated from the tachograms of the circadian profiles of 24 hours every 5 minutes (288 frames) of each subject, and part of this data were used to train the network. The classification work done by the deep neural network had 98% of accuracy and 98% of precision, validated with the ROC curve, whose AUC values were approximately the unit for each group. Taking into account the good performance, we can consider this deep neural network and approximate entropy as useful tools to have a good early diagnosis about Chagas disease and its cardiac compromise.
{"title":"Approximate Entropy and Densely Connected Neural Network in the Early Diagnostic of Patients with Chagas Disease","authors":"María Fernanda Rodríguez, A. Ravelo-García, E. Alvarez, Luz Alexandra Díaz, D. Cornejo, Victor Cabrera-Caso, Dante Condori-Merma, Miguel Vizcardo Cornejo","doi":"10.22489/CinC.2022.313","DOIUrl":"https://doi.org/10.22489/CinC.2022.313","url":null,"abstract":"It is estimated that in the world there are between 6 and 8 million people infected with Chagas disease, mainly in endemic areas of 21 Latin American countries, and in recent years it is slowly becoming a health problem in more urban areas and countries. In that sense, developing diagnosis methods is primordial. That is why this work used a deep neural network to classify 292 subjects (volunteers and patients) composed of 83 health volunteers (Control group); 102 asymptomatic chagasic patients (CH1 group) and 107 seropositive chagasic patients with incipient heart disease (CH2 group). Approximate Entropy ApEn was calculated from the tachograms of the circadian profiles of 24 hours every 5 minutes (288 frames) of each subject, and part of this data were used to train the network. The classification work done by the deep neural network had 98% of accuracy and 98% of precision, validated with the ROC curve, whose AUC values were approximately the unit for each group. Taking into account the good performance, we can consider this deep neural network and approximate entropy as useful tools to have a good early diagnosis about Chagas disease and its cardiac compromise.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"24 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115686382","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
B. Cairo, V. Bari, F. Gelpi, Beatrice De Maria, Anita Mollo, F. Bandera, A. Porta
Evaluation of cardiorespiratory coupling (CRC) usually requires the simultaneous recording of heart period (HP) variability, derived from the electrocardiogram (ECG), and respiration. ECG-derived respiration (ECGDR) exploits the cardiac axis movement due to respiration to estimate respiratory activity directly from the ECG. Since CRC indexes could theoretically be computed using ECGDR, a comparison with results obtained through a more precise monitoring of respiratory activity such as the respiratory flow (RF) is warranted. Therefore, a mixed unpredictability index (MUPI) of HP variability from respiratory dynamics, computed via local k-nearest-neighbor approach, was calculated using ECGDR and RF in patients with preserved functional capacity (PFC) and with reduced functional capacity (RFC) before and after cardiopulmonary exercise test (CPET) protocol. The MUPI computed from RF was found to be significantly increased in PFC patients after CPET protocol, while no effect could be observed when considering the ECGDR. Moreover, the correlation between the two MUPI indexes was limited. We conclude that indexes of CRC might require more direct measures of respiration than ECGDR to detect pathophysiological differences.
{"title":"Comparison Between ECG-Derived Respiration and Respiratory Flow for the Assessment of Cardiorespiratory Coupling Before and After Cardiopulmonary Exercise Test Protocol","authors":"B. Cairo, V. Bari, F. Gelpi, Beatrice De Maria, Anita Mollo, F. Bandera, A. Porta","doi":"10.22489/CinC.2022.103","DOIUrl":"https://doi.org/10.22489/CinC.2022.103","url":null,"abstract":"Evaluation of cardiorespiratory coupling (CRC) usually requires the simultaneous recording of heart period (HP) variability, derived from the electrocardiogram (ECG), and respiration. ECG-derived respiration (ECGDR) exploits the cardiac axis movement due to respiration to estimate respiratory activity directly from the ECG. Since CRC indexes could theoretically be computed using ECGDR, a comparison with results obtained through a more precise monitoring of respiratory activity such as the respiratory flow (RF) is warranted. Therefore, a mixed unpredictability index (MUPI) of HP variability from respiratory dynamics, computed via local k-nearest-neighbor approach, was calculated using ECGDR and RF in patients with preserved functional capacity (PFC) and with reduced functional capacity (RFC) before and after cardiopulmonary exercise test (CPET) protocol. The MUPI computed from RF was found to be significantly increased in PFC patients after CPET protocol, while no effect could be observed when considering the ECGDR. Moreover, the correlation between the two MUPI indexes was limited. We conclude that indexes of CRC might require more direct measures of respiration than ECGDR to detect pathophysiological differences.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114595141","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Alexandra Jamieson, M. Orini, N. Chaturvedi, Alun D. Hughes
We determined wearable device errors in assessing a 6-Minute Walk Test (6MWT). 16 healthy adults (male 7(44%), mean $agepm SD 27pm 4$ years) performed a standard (6MWT-S) and modified, free range’, (6MWT-FR) protocols with a Garmin and Fitbit smartwatch to measure three parameters: distance, step count and heart rate (HR). Distance during the 6MWT-FR was measured with smaller errors during 6MWT-S for both Garmin (Mean Absolute Percentage Error, $MAPE=9.8{%}$ [4.6%,12.6%] $vs quad 18.5%[13.0%,27.4%],p < 0.001)$ and Fitbit $(M A P E=9.4 %[4.5 %, 13.3 %] {vs } 22.7 %[18.3 %, 29.3 %],p < 0.001)$. Steps were measured with smaller errors with Garmin $(M A P E=2.3 %[1.1 %, 2.9 %]; r=0.96)$ than Fitbit (Fitbit: $MAPE=8.1%[5.0%,12.9%]; r=0.24)$. Heart rate at rest, peak exercise and recovery was measured with median MAPE ranging between 1.2% and $2.9{%}$, with no evidence of difference between the two devices. Wearable measurements of the 6MWT provide insights about exercise capacity which could be monitored and evaluated remotely.
我们在评估6分钟步行测试(6MWT)时确定了可穿戴设备的误差。16名健康成年人(男性7人(44%)，平均年龄27岁，4岁)使用Garmin和Fitbit智能手表进行标准(6MWT-S)和改良的自由放养(6MWT-FR)方案，测量三个参数:距离、步数和心率(HR)。Garmin(平均绝对百分比误差，$MAPE=9.8{%}$ [4.6%，12.6%] $vs quad 18.5%[13.0%，27.4%]，p < 0.001)$和Fitbit $(m.a p E=9.4 %[4.5 %， 13.3 %] {对 22.7 %[18.3 %，29.3 %]，p < 0.001)$在6MWT-S期间测量的距离误差较小。用Garmin $(M A P E=2.3 %[1.1 %， 2.9 %])测量步数误差较小;r = 0.96)比美元Fitbit (Fitbit:日军= 8.1美元 % (12.9 5.0 %,%);美元 r = 0.24)。在休息、运动高峰和恢复时的心率测量中位数MAPE在1.2%到2.9之间，没有证据表明两种设备之间存在差异。6MWT的可穿戴测量提供了关于运动能力的见解，可以远程监测和评估。
{"title":"A Validation Study of Two Wrist Worn Wearable Devices for Remote Assessment of Exercise Capacity","authors":"Alexandra Jamieson, M. Orini, N. Chaturvedi, Alun D. Hughes","doi":"10.22489/CinC.2022.259","DOIUrl":"https://doi.org/10.22489/CinC.2022.259","url":null,"abstract":"We determined wearable device errors in assessing a 6-Minute Walk Test (6MWT). 16 healthy adults (male 7(44%), mean $agepm SD 27pm 4$ years) performed a standard (6MWT-S) and modified, free range’, (6MWT-FR) protocols with a Garmin and Fitbit smartwatch to measure three parameters: distance, step count and heart rate (HR). Distance during the 6MWT-FR was measured with smaller errors during 6MWT-S for both Garmin (Mean Absolute Percentage Error, $MAPE=9.8{%}$ [4.6%,12.6%] $vs quad 18.5%[13.0%,27.4%],p < 0.001)$ and Fitbit $(M A P E=9.4 %[4.5 %, 13.3 %] {vs } 22.7 %[18.3 %, 29.3 %],p < 0.001)$. Steps were measured with smaller errors with Garmin $(M A P E=2.3 %[1.1 %, 2.9 %]; r=0.96)$ than Fitbit (Fitbit: $MAPE=8.1%[5.0%,12.9%]; r=0.24)$. Heart rate at rest, peak exercise and recovery was measured with median MAPE ranging between 1.2% and $2.9{%}$, with no evidence of difference between the two devices. Wearable measurements of the 6MWT provide insights about exercise capacity which could be monitored and evaluated remotely.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123082412","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
S. J. Hill, Alistair A. Young, R. Rajani, A. Vecchi
Transcatheter Valve Embolization and Migration (TVEM) is a rare, but catastrophic event where the prosthesis moves due to heamodynamic forces acting on the frame. TVEM following Transcatheter Mitral Valve Replacement (TMVR) is largely undocumented. Haemodynamic forces cannot be estimated during pre-procedural planning and conventional imaging does not allow to compute them after replacement. To shed light on this issue, this study focusses on modelling haemodynamics after TMVR in 3 patients with Mitral Annular Calcification (MAC) known as Valve-in-MAC (ViMAC). Three-dimensional flow simulations are performed using the computational fluid dynamics (CFD) package STARCCM+. Results of the simulation are processed to compute the fluid forces acting on the device and pressure gradients in the left ventricular outflow tract (LVOT). Anatomical measurements are performed on CT data sets to assess the mitral valve size and shape, the extent and location of the calcification and the size of the LVOT after implantation. Our results show that the force distribution on the device is largely influenced by LVOT anatomy and contraction patterns.
{"title":"Assessment of Transcatheter Heart Valve Migration and Embolization Risk Following Valve-in-MAC","authors":"S. J. Hill, Alistair A. Young, R. Rajani, A. Vecchi","doi":"10.22489/CinC.2022.428","DOIUrl":"https://doi.org/10.22489/CinC.2022.428","url":null,"abstract":"Transcatheter Valve Embolization and Migration (TVEM) is a rare, but catastrophic event where the prosthesis moves due to heamodynamic forces acting on the frame. TVEM following Transcatheter Mitral Valve Replacement (TMVR) is largely undocumented. Haemodynamic forces cannot be estimated during pre-procedural planning and conventional imaging does not allow to compute them after replacement. To shed light on this issue, this study focusses on modelling haemodynamics after TMVR in 3 patients with Mitral Annular Calcification (MAC) known as Valve-in-MAC (ViMAC). Three-dimensional flow simulations are performed using the computational fluid dynamics (CFD) package STARCCM+. Results of the simulation are processed to compute the fluid forces acting on the device and pressure gradients in the left ventricular outflow tract (LVOT). Anatomical measurements are performed on CT data sets to assess the mitral valve size and shape, the extent and location of the calcification and the size of the LVOT after implantation. Our results show that the force distribution on the device is largely influenced by LVOT anatomy and contraction patterns.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"107 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123245906","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
D. Cornejo, A. Ravelo-García, E. Alvarez, María Fernanda Rodríguez, Luz Alexandra Díaz, Victor Cabrera-Caso, Dante Condori-Merma, Miguel Vizcardo Cornejo
Chagas disease is a life threatening illness that in the last decades was becoming a public health problem because of the change in the epidemiological pattern. It may be silent and asymptomatic in the chronic phase. Hence the necessity of the development of early markers. To achieve this, we propose a deep neural network architecture in order to classify 292 patients into three groups: The Control group with 83 volunteers, the CH1 group with 102 patients with positive serology and no cardiac involvement and the CH2 group with 107 patients with positive serology and incipient heart failure. The used data comes from 24-hour ECG, the RR intervals from each subject was divided in 288 frames of 5 minutes each. Then it was preprocessed using permutation entropy obtaining the circadian profile for each patient. And by applying PCA each patient ended up represented by a vector of 144 entries. This was in turn used for the training of the proposed NN architecture. The classification performed with 91% accuracy and an average of 92% precision, consisting in a great work of classification validated by the AUC in each ROC curve. As this results were obtained with a limited quantity of data, this study can be improved provided with more samples, making this model a tool for analyzing ECG in order to try to do an early evaluation and diagnosis of a cardiac compromise related to the generally silent chronic phase.
{"title":"Deep Learning and Permutation Entropy in the Stratification of Patients with Chagas Disease","authors":"D. Cornejo, A. Ravelo-García, E. Alvarez, María Fernanda Rodríguez, Luz Alexandra Díaz, Victor Cabrera-Caso, Dante Condori-Merma, Miguel Vizcardo Cornejo","doi":"10.22489/CinC.2022.311","DOIUrl":"https://doi.org/10.22489/CinC.2022.311","url":null,"abstract":"Chagas disease is a life threatening illness that in the last decades was becoming a public health problem because of the change in the epidemiological pattern. It may be silent and asymptomatic in the chronic phase. Hence the necessity of the development of early markers. To achieve this, we propose a deep neural network architecture in order to classify 292 patients into three groups: The Control group with 83 volunteers, the CH1 group with 102 patients with positive serology and no cardiac involvement and the CH2 group with 107 patients with positive serology and incipient heart failure. The used data comes from 24-hour ECG, the RR intervals from each subject was divided in 288 frames of 5 minutes each. Then it was preprocessed using permutation entropy obtaining the circadian profile for each patient. And by applying PCA each patient ended up represented by a vector of 144 entries. This was in turn used for the training of the proposed NN architecture. The classification performed with 91% accuracy and an average of 92% precision, consisting in a great work of classification validated by the AUC in each ROC curve. As this results were obtained with a limited quantity of data, this study can be improved provided with more samples, making this model a tool for analyzing ECG in order to try to do an early evaluation and diagnosis of a cardiac compromise related to the generally silent chronic phase.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123427664","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Electrocardiography is well established as an effective clinical tool for detection and diagnosis of cardiac arrhythmias and abnormalities. The objective of the 2021 PhysioNet/Computing in Cardiology Challenge was for teams to develop automated classification algorithms for reduced-lead ECGs. While it is well-known that proper pre-processing is very important for the success of classification algorithms, there is not universal agreement as to the appropriate pre-processing steps for automated ECG classification. Papers from the top 15 finishers in the Challenge as well as the bottom ten finishers were examined to determine what pre-processing steps were applied by each team. The most commonly used pre-processing steps included resampling to a consistent sampling rate, applying a bandpass filter, normalizing and using a fixed signal length. There were a number of similarities in the preprocessing steps used by the top 15 teams, whereas all of these steps were not applied in the majority of approaches for the bottom ten teams. In the bottom ten participants, less than half used a bandpass filter, and only three applied some type of normalization. This investigation underscores the importance of appropriate pre-processing for strong classification accuracy and the need for a universal approach to pre-processing techniques in automated ECG classification.
心电图是一种有效的检测和诊断心律失常和异常的临床工具。2021年PhysioNet/Computing in Cardiology挑战赛的目标是让团队开发用于减少导联心电图的自动分类算法。虽然众所周知，适当的预处理对于分类算法的成功是非常重要的，但对于自动心电分类的适当预处理步骤并没有普遍的共识。来自挑战赛前15名和后10名的论文将被检查，以确定每个团队应用了哪些预处理步骤。最常用的预处理步骤包括重新采样到一致的采样率，应用带通滤波器，归一化和使用固定的信号长度。在前15个团队使用的预处理步骤中有许多相似之处，而所有这些步骤并没有应用于后10个团队的大多数方法中。在最后十位参与者中，不到一半的人使用了带通滤波器，只有三个人应用了某种类型的归一化。这项研究强调了适当的预处理对强分类准确性的重要性，以及在自动心电分类中需要一种通用的预处理技术。
{"title":"Impact of Pre-Processing Decisions on Automated ECG Classification Accuracy","authors":"Adrian K. Cornely, Grace M. Mirsky","doi":"10.22489/CinC.2022.252","DOIUrl":"https://doi.org/10.22489/CinC.2022.252","url":null,"abstract":"Electrocardiography is well established as an effective clinical tool for detection and diagnosis of cardiac arrhythmias and abnormalities. The objective of the 2021 PhysioNet/Computing in Cardiology Challenge was for teams to develop automated classification algorithms for reduced-lead ECGs. While it is well-known that proper pre-processing is very important for the success of classification algorithms, there is not universal agreement as to the appropriate pre-processing steps for automated ECG classification. Papers from the top 15 finishers in the Challenge as well as the bottom ten finishers were examined to determine what pre-processing steps were applied by each team. The most commonly used pre-processing steps included resampling to a consistent sampling rate, applying a bandpass filter, normalizing and using a fixed signal length. There were a number of similarities in the preprocessing steps used by the top 15 teams, whereas all of these steps were not applied in the majority of approaches for the bottom ten teams. In the bottom ten participants, less than half used a bandpass filter, and only three applied some type of normalization. This investigation underscores the importance of appropriate pre-processing for strong classification accuracy and the need for a universal approach to pre-processing techniques in automated ECG classification.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"97 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124724460","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
T. Zlahtic, D. Žižek, M. Mrak, A. Z. Mežnar, V. Starc
Cardiac resynchronization therapy with biventricular pacing (BiV) is the cornerstone treatment for heart failure patients with ventricular dyssynchrony. Recently, the conduction system pacing (CSP) has being introduced as a possible alternative. We hypothesized that CSP could produce a more complete electrical resynchronization compared to conventional BIV pacing. To trace the spreading of myocardial depolarization, we assessed equivalent dipole (ED) trajectories utilizing the BEM method with a tailored human torso from the high resolution 12-lead ECG before and after device implantation in 17 patients included in our ongoing randomized CSP-SYNC study. We observed a similar relative shortening of the QRS duration (0,23 in CSP and 0,25 in BiV) and relative ED trajectory length (0,16 in CSP and 0,20 in BiV). However, a significant change of ED trajectory direction occurred after the therapy. In BiV pacing, the trajectory direction shifted more towards the base of the heart, but more apically in CSP, mimicking normal heart depolarization Resynchronization with CSP seems to restore more physiological depolarization compared to BiV pacing. The assessment of the ED trajectories provides additional insight into the electrical heart remodelling after the therapy.
{"title":"Conduction System Pacing Versus Biventricular Pacing For Cardiac Resynchronization - Preliminary Electrocardiographic Results","authors":"T. Zlahtic, D. Žižek, M. Mrak, A. Z. Mežnar, V. Starc","doi":"10.22489/CinC.2022.297","DOIUrl":"https://doi.org/10.22489/CinC.2022.297","url":null,"abstract":"Cardiac resynchronization therapy with biventricular pacing (BiV) is the cornerstone treatment for heart failure patients with ventricular dyssynchrony. Recently, the conduction system pacing (CSP) has being introduced as a possible alternative. We hypothesized that CSP could produce a more complete electrical resynchronization compared to conventional BIV pacing. To trace the spreading of myocardial depolarization, we assessed equivalent dipole (ED) trajectories utilizing the BEM method with a tailored human torso from the high resolution 12-lead ECG before and after device implantation in 17 patients included in our ongoing randomized CSP-SYNC study. We observed a similar relative shortening of the QRS duration (0,23 in CSP and 0,25 in BiV) and relative ED trajectory length (0,16 in CSP and 0,20 in BiV). However, a significant change of ED trajectory direction occurred after the therapy. In BiV pacing, the trajectory direction shifted more towards the base of the heart, but more apically in CSP, mimicking normal heart depolarization Resynchronization with CSP seems to restore more physiological depolarization compared to BiV pacing. The assessment of the ED trajectories provides additional insight into the electrical heart remodelling after the therapy.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"8 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121708496","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Miriam Goldammer, S. Zaunseder, Franz Ehrlich, Hagen Malberg
This work investigates the benefit of using multiple signals and preprocessing strategies for sleep staging from cardiorespiratory signals. We modified our previous Neural Network model to take different signal combinations as input. To that end, we added oxygen saturation and different respiratory signals to the electrocardiogram. We further invoked different preprocessing strategies that have been described previously for such signals, namely using downsampled signals vs. using time series of breath-to-breath intervals. We trained and tested our model variations with 4784 polysomnograms from the Sleep Heart Health Study. We found the best combination of signals to be heart rate together with a downsampled respiratory signal. The classification resulted in a k of 0.68 on hold-out test data, which outperforms our previous results and state of the art for cardiorespiratory sleep staging. We observe that combinations of cardiorespiratory signals can improve classification performance for automatic cardiorespiratory sleep staging. As there are generally more cardiorespiratory signals available and many more options for preprocessing them, we expect that further research in this area will show even more improvements.
{"title":"Comparison of Signal Combinations for Cardiorespiratory Sleep Staging","authors":"Miriam Goldammer, S. Zaunseder, Franz Ehrlich, Hagen Malberg","doi":"10.22489/CinC.2022.077","DOIUrl":"https://doi.org/10.22489/CinC.2022.077","url":null,"abstract":"This work investigates the benefit of using multiple signals and preprocessing strategies for sleep staging from cardiorespiratory signals. We modified our previous Neural Network model to take different signal combinations as input. To that end, we added oxygen saturation and different respiratory signals to the electrocardiogram. We further invoked different preprocessing strategies that have been described previously for such signals, namely using downsampled signals vs. using time series of breath-to-breath intervals. We trained and tested our model variations with 4784 polysomnograms from the Sleep Heart Health Study. We found the best combination of signals to be heart rate together with a downsampled respiratory signal. The classification resulted in a k of 0.68 on hold-out test data, which outperforms our previous results and state of the art for cardiorespiratory sleep staging. We observe that combinations of cardiorespiratory signals can improve classification performance for automatic cardiorespiratory sleep staging. As there are generally more cardiorespiratory signals available and many more options for preprocessing them, we expect that further research in this area will show even more improvements.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"260 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124256043","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The normal electrical potential propagates throughout the atria periodically. During atrial arrhythmias its prop-agation is modified because the substrate is not homoge-neous and new sources of punctual electrical activity appear. The periodic behavior of activation remains predom-inant, but becomes local in different parts of the atria. It is characterized by cycle length (CL) which measures the frequency of activation and can be computed from intrac-ardiac bipolar electrograms (EGM) recorded by a mapping catheter during the catheter ablation procedure. The CL value of different mapped zones is an extremely important resource for physicians when performing persis-tent Atrial Fibrillation (AF) ablation because it helps to identify pathological zones and define the ablation strat-egy. Thus, a reliable estimation of the CL of atrial tissue is essential. The complexity of this task stems from the large variability in EGM morphology influenced by mul-tiple wavefronts, fragmentation and added noise. In this work, we propose a cycle length estimator that can process the complex mapping signals recorded during atrial arrhythmias ablation and reliably provide the frequency of their periodic activity.
{"title":"Cycle Length Estimation Using Accurate Adaptive Detection of Local Activations in Atrial Intracardiac Electrograms","authors":"Dinara Veshchezerova, C. Bars, J. Seitz","doi":"10.22489/CinC.2022.142","DOIUrl":"https://doi.org/10.22489/CinC.2022.142","url":null,"abstract":"The normal electrical potential propagates throughout the atria periodically. During atrial arrhythmias its prop-agation is modified because the substrate is not homoge-neous and new sources of punctual electrical activity appear. The periodic behavior of activation remains predom-inant, but becomes local in different parts of the atria. It is characterized by cycle length (CL) which measures the frequency of activation and can be computed from intrac-ardiac bipolar electrograms (EGM) recorded by a mapping catheter during the catheter ablation procedure. The CL value of different mapped zones is an extremely important resource for physicians when performing persis-tent Atrial Fibrillation (AF) ablation because it helps to identify pathological zones and define the ablation strat-egy. Thus, a reliable estimation of the CL of atrial tissue is essential. The complexity of this task stems from the large variability in EGM morphology influenced by mul-tiple wavefronts, fragmentation and added noise. In this work, we propose a cycle length estimator that can process the complex mapping signals recorded during atrial arrhythmias ablation and reliably provide the frequency of their periodic activity.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125435098","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Jiayi Yan, Hanshuang Xie, Huaiyu Zhu, Yamin Liu, Fan Wu, Yun Pan
Accurate P wave detection is important for arrhythmia diagnosis, e.g. P wave absence or P duration for atrial fibrillation diagnosis and other atrial arrhythmias. Phasor transform is an effective method for ECG fiducial points delineation. It maps each ECG sample into a phasor to enhance slight variations and preserves morphology and magnitude characteristics. In this paper, we optimized the automatic P wave delineation method based on phasor transform in four aspects, i.e., signal denoising, wave localization, candidate points detection, and optimal points selection. In our experiments, the length of the search window and the degree of phasor transform were established through various trials. Especially, along with zero-crossing points of the phasor signal, intersections of the phasor signal and the original ECG signal are obtained as candidates, which make the most contribution to delineation results. For validation, the QT Database with 3194 P wave annotations from 105 records of two leads is adopted. As a result, we reached F1 scores of 94.67% and 93.56% with detection error rates (DERs) of 10.80% and 13.06% for P wave onset and offset points detection, respectively. The F 1 score and DER for P peak detection under a tolerance of 75 ms were 95.33% and 9.46%, respectively, which outperforms other reproducible works and their combinations.
{"title":"An Optimized Automatic P Wave Delineation Method Based on Phasor Transform","authors":"Jiayi Yan, Hanshuang Xie, Huaiyu Zhu, Yamin Liu, Fan Wu, Yun Pan","doi":"10.22489/CinC.2022.122","DOIUrl":"https://doi.org/10.22489/CinC.2022.122","url":null,"abstract":"Accurate P wave detection is important for arrhythmia diagnosis, e.g. P wave absence or P duration for atrial fibrillation diagnosis and other atrial arrhythmias. Phasor transform is an effective method for ECG fiducial points delineation. It maps each ECG sample into a phasor to enhance slight variations and preserves morphology and magnitude characteristics. In this paper, we optimized the automatic P wave delineation method based on phasor transform in four aspects, i.e., signal denoising, wave localization, candidate points detection, and optimal points selection. In our experiments, the length of the search window and the degree of phasor transform were established through various trials. Especially, along with zero-crossing points of the phasor signal, intersections of the phasor signal and the original ECG signal are obtained as candidates, which make the most contribution to delineation results. For validation, the QT Database with 3194 P wave annotations from 105 records of two leads is adopted. As a result, we reached F1 scores of 94.67% and 93.56% with detection error rates (DERs) of 10.80% and 13.06% for P wave onset and offset points detection, respectively. The F 1 score and DER for P peak detection under a tolerance of 75 ms were 95.33% and 9.46%, respectively, which outperforms other reproducible works and their combinations.","PeriodicalId":117840,"journal":{"name":"2022 Computing in Cardiology (CinC)","volume":"4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126512012","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：