Considerations on Performance Evaluation of Atrial Fibrillation Detectors

Butkuvienė, Monika; Petrėnas, Andrius; Sološenko, Andrius; Martín-Yebra, Alba; Marozas, Vaidotas; Sörnmo, Leif

doi:10.1109/tbme.2021.3067698

Cited by 28 publications

(12 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, when the classes in the datasets are unbalanced (i.e., the number of elements in one class is much larger than the number of elements in the other class), Acc and F 1 scores tend to be inflated. Therefore, in order to address this issue, it has been suggested to report the Mcc score as it does not seem to be influenced by unbalanced classes (Chicco and Jurman, 2020 ; Butkuviene et al, 2021 ). Additionally, F 1 score has the disadvantages that it varies when swapping the classes, and that it is independent from TN predictions (Chicco and Jurman, 2020 ).…”

Section: Discussionmentioning

confidence: 99%

Detection of Brief Episodes of Atrial Fibrillation Based on Electrocardiomatrix and Convolutional Neural Network

Martinez

Bie²,

Marzocchi³

et al. 2021

Front. Physiol.

View full text Add to dashboard Cite

Background: Brief episodes of atrial fibrillation (AF) may evolve into longer AF episodes increasing the chances of thrombus formation, stroke, and death. Classical methods for AF detection investigate rhythm irregularity or P-wave absence in the ECG, while deep learning approaches profit from the availability of annotated ECG databases to learn discriminatory features linked to different diagnosis. However, some deep learning approaches do not provide analysis of the features used for classification. This paper introduces a convolutional neural network (CNN) approach for automatic detection of brief AF episodes based on electrocardiomatrix-images (ECM-images) aiming to link deep learning to features with clinical meaning.Materials and Methods: The CNN is trained using two databases: the Long-Term Atrial Fibrillation and the MIT-BIH Normal Sinus Rhythm, and tested on three databases: the MIT-BIH Atrial Fibrillation, the MIT-BIH Arrhythmia, and the Monzino-AF. Detection of AF is done using a sliding window of 10 beats plus 3 s. Performance is quantified using both standard classification metrics and the EC57 standard for arrhythmia detection. Layer-wise relevance propagation analysis was applied to link the decisions made by the CNN to clinical characteristics in the ECG.Results: For all three testing databases, episode sensitivity was greater than 80.22, 89.66, and 97.45% for AF episodes shorter than 15, 30 s, and for all episodes, respectively.Conclusions: Rhythm and morphological characteristics of the electrocardiogram can be learned by a CNN from ECM-images for the detection of brief episodes of AF.

show abstract

Section: Discussionmentioning

confidence: 99%

Detection of Brief Episodes of Atrial Fibrillation Based on Electrocardiomatrix and Convolutional Neural Network

Martinez

Bie²,

Marzocchi³

et al. 2021

Front. Physiol.

View full text Add to dashboard Cite

show abstract

“…For each threshold x with the optimal cutoff value for differentiating between AF and SR, we calculated several classification metrics [ 41 ], namely, accuracy, specificity, sensitivity, F1-score [ 42 ], positive predictive value (PPV), negative predictive value (NPV), and diagnostic odds ratio (DOR) [ 43 ]. For the estimation of classification metrics’ 95% confidence interval (CI), we used a nonparametric bootstrap with 5000 samples [ 44 ].…”

Section: Materials and Methodsmentioning

confidence: 99%

Statistical and Diagnostic Properties of pRRx Parameters in Atrial Fibrillation Detection

Buś

Jędrzejewski

Guzik

2022

JCM

View full text Add to dashboard Cite

Background: We studied the diagnostic properties of the percentage of successive RR intervals differing by at least x ms (pRRx) as functions of the threshold value x in a range of 7 to 195 ms for the differentiation of atrial fibrillation (AF) from sinus rhythm (SR). Methods: RR intervals were measured in 60-s electrocardiogram (ECG) segments with either AF (32,141 segments) or SR (32,769 segments) from the publicly available Physionet Long-Term Atrial Fibrillation Database (LTAFDB). For validation, we have used ECGs from the Massachusetts Institute of Technology–Beth Israel Hospital (MIT–BIH) Atrial Fibrillation Database. The pRRx distributions in AF and SR in relation to x were studied by histograms, along with the mutual association by the nonparametric Spearman correlations for all pairs of pRRx, and separately for AF or SR. The optimal cutoff values for all pRRx were determined using the receiver operator curve characteristic. A nonparametric bootstrap with 5000 samples was used to calculate a 95% confidence interval for several classification metrics. Results: The distributions of pRRx for x in the 7–195 ms range are significantly different in AF than in SR. The sensitivity, specificity, accuracy, and diagnostic odds ratios differ for pRRx, with the highest values for x = 31 ms (pRR31) rather than x = 50 (pRR50), which is most commonly applied in studies on heart rate variability. For the optimal cutoff of pRR31 (68.79%), the sensitivity is 90.42%, specificity 95.37%, and the diagnostic odds ratio is 194.11. Validation with the ECGs from the MIT–BIH Atrial Fibrillation Database confirmed our findings. Conclusions: We demonstrate that the diagnostic properties of pRRx depend on x, and pRR31 outperforms pRR50, at least for ECGs of 60-s duration.

show abstract

“…Some methods [9], [32], [33] applied machine learning methods such as support vector machine (SVM) [32], convolutional neural network [33] or neighborhood component analysis [9] to R-R intervals for AF detection and achieved higher performance than traditional methods. However, it is difficult for rhythm-based methods using R-R intervals to distinguish AF from other arrhythmias, while the morphology-based methods such as P-wave absence can avoid false-positive errors in AF detection [34]. Deep learningbased methods [10]- [12] can be directly applied to ECG signals and achieve good AF detection performance with large datasets, but they are sensitive to changes in ECG morphology and are unclear how they generalize to unseen data [34].…”

Section: A Contact Methods For Af Detectionmentioning

confidence: 99%

“…However, it is difficult for rhythm-based methods using R-R intervals to distinguish AF from other arrhythmias, while the morphology-based methods such as P-wave absence can avoid false-positive errors in AF detection [34]. Deep learningbased methods [10]- [12] can be directly applied to ECG signals and achieve good AF detection performance with large datasets, but they are sensitive to changes in ECG morphology and are unclear how they generalize to unseen data [34].…”

Section: A Contact Methods For Af Detectionmentioning

confidence: 99%

Non-Contact Atrial Fibrillation Detection From Face Videos by Learning Systolic Peaks

Sun

Junttila

Tulppo

et al. 2022

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

We propose a non-contact approach for atrial fibrillation (AF) detection from face videos. Methods: Face videos, electrocardiography (ECG), and contact photoplethysmography (PPG) from 100 healthy subjects and 100 AF patients are recorded. Data recordings from healthy subjects are all labeled as healthy. Two cardiologists evaluated ECG recordings of patients and labeled each recording as AF, sinus rhythm (SR), or atrial flutter (AFL). We use the 3D convolutional neural network for remote PPG monitoring and propose a novel loss function (Wasserstein distance) to use the timing of systolic peaks from contact PPG as the label for our model training. Then a set of heart rate variability (HRV) features are calculated from the inter-beat intervals, and a support vector machine (SVM) classifier is trained with HRV features. Results: Our proposed method can accurately extract systolic peaks from face videos for AF detection. The proposed method is trained with subject-independent 10-fold cross-validation with 30 s video clips and tested on two tasks. 1) Classification of healthy versus AF: the accuracy, sensitivity, and specificity are 96.00%, 95.36%, and 96.12%. 2) Classification of SR versus AF: the accuracy, sensitivity, and specificity are 95.23%, 98.53%, and 91.12%. In addition, we also demonstrate the feasibility of non-contact AFL detection. Conclusion: We achieve good performance of non-contact AF detection by learning systolic peaks. Significance: noncontact AF detection can be used for self-screening of AF symptoms for suspectable populations at home or selfmonitoring of AF recurrence after treatment for chronic patients.

show abstract

Considerations on Performance Evaluation of Atrial Fibrillation Detectors

Cited by 28 publications

References 71 publications

Detection of Brief Episodes of Atrial Fibrillation Based on Electrocardiomatrix and Convolutional Neural Network

Detection of Brief Episodes of Atrial Fibrillation Based on Electrocardiomatrix and Convolutional Neural Network

Statistical and Diagnostic Properties of pRRx Parameters in Atrial Fibrillation Detection

Non-Contact Atrial Fibrillation Detection From Face Videos by Learning Systolic Peaks

Contact Info

Product

Resources

About