Contactless facial video recording with deep learning models for the detection of atrial fibrillation.
Ontology highlight
ABSTRACT:
SUBMITTER: Sun Y
PROVIDER: S-EPMC8741942 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:BackgroundAtrial fibrillation (AF), a common cause of stroke, often is asymptomatic. Smartphones and smartwatches can detect AF using heart rate patterns inferred using photoplethysmography (PPG); however, enhanced accuracy is required to reduce false positives in screening populations.ObjectiveThe purpose of this study was to test the hypothesis that a deep learning algorithm given raw, smartwatch-derived PPG waveforms would discriminate AF from normal sinus rhythm better than algorithms using heart rate alone.MethodsPatients presenting for cardioversion of AF (n = 51) were given wrist-worn fitness trackers containing PPG sensors (Jawbone Health). Standard 12-lead electrocardiograms over-read by board-certified cardiac electrophysiologists were used as the reference standard. The accuracy of PPG signals to discriminate AF from sinus rhythm was evaluated by conventional measures of heart rate variability, a long short-term memory (LSTM) neural network given heart rate data only, and a deep convolutional-recurrent neural net (DNN) given the raw PPG data.ResultsFrom among 51 patients with persistent AF (age 63.6 ± 11.3 years; 78% male; 88% white), we randomly assigned 40 to train and 11 to test the algorithms. Whereas logistic regression analysis of heart rate variability yielded an area under the receiver operating characteristic curve (AUC) of 0.717 (sensitivity 0.741; specificity 0.584), the LSTM model given heart rate data exhibited AUC of 0.954 (sensitivity 0.810; specificity 0.921), and the DNN model given raw PPG data yielded the highest AUC of 0.983 (sensitivity 0.985; specificity 0.880).ConclusionA deep learning model given the raw PPG-based signal resulted in AF detection with high accuracy, performing better than conventional analyses relying on heart rate series data alone.
| S-EPMC8183963 | biostudies-literature
Project description:Objective and Impact Statement. Atrial fibrillation (AF) is a serious medical condition that requires effective and timely treatment to prevent stroke. We explore deep neural networks (DNNs) for learning cardiac cycles and reliably detecting AF from single-lead electrocardiogram (ECG) signals. Introduction. Electrocardiograms are widely used for diagnosis of various cardiac dysfunctions including AF. The huge amount of collected ECGs and recent algorithmic advances to process time-series data with DNNs substantially improve the accuracy of the AF diagnosis. DNNs, however, are often designed as general purpose black-box models and lack interpretability of their decisions. Methods. We design a three-step pipeline for AF detection from ECGs. First, a recording is split into a sequence of individual heartbeats based on R-peak detection. Individual heartbeats are then encoded using a DNN that extracts interpretable features of a heartbeat by disentangling the duration of a heartbeat from its shape. Second, the sequence of heartbeat codes is passed to a DNN to combine a signal-level representation capturing heart rhythm. Third, the signal representations are passed to a DNN for detecting AF. Results. Our approach demonstrates a superior performance to existing ECG analysis methods on AF detection. Additionally, the method provides interpretations of the features extracted from heartbeats by DNNs and enables cardiologists to study ECGs in terms of the shapes of individual heartbeats and rhythm of the whole signals. Conclusion. By considering ECGs on two levels and employing DNNs for modelling of cardiac cycles, this work presents a method for reliable detection of AF from single-lead ECGs.
| S-EPMC10521743 | biostudies-literature
Project description:PurposeTo accurately record the movements of a hand-held target together with the smooth pursuit eye movements (SPEMs) elicited with video-oculography (VOG) combined with deep learning-based object detection using a single-shot multibox detector (SSD).MethodsThe SPEMs of 11 healthy volunteers (21.3 ± 0.9 years) were recorded using VOG. The subjects fixated on a moving target that was manually moved at a distance of 1 m by the examiner. An automatic recording system was developed using SSD to predict the type and location of objects in a single image. The 400 images that were taken of one subject using a VOG scene camera were distributed into 2 groups (300 and 100) for training and validation. The testing data included 1100 images of all subjects (100 images/subject). The method achieved 75% average precision (AP75) for the relationship between the location of the fixated target (as calculated by SSD) and the position of each eye (as recorded by VOG).ResultsThe AP75 for all subjects was 99.7% ± 0.6%. The horizontal and vertical target locations were significantly and positively correlated with each eye position in the horizontal and vertical directions (adjusted R2 ≥ 0.955, P < 0.001).ConclusionsThe addition of SSD-driven recording of hand-held target positions with VOG allows for quantitative assessment of SPEMs following a target during an SPEM test.Translational relevanceThe combined methods of VOG and SSD can be used to detect SPEMs with greater accuracy, which can improve the outcome of clinical evaluations.
| S-EPMC8107482 | biostudies-literature
Project description:BACKGROUND:Atrial Fibrillation is the most common arrhythmia worldwide with a global age adjusted prevalence of 0.5% in 2010. Anticoagulation treatment using warfarin or direct oral anticoagulants is effective in reducing the risk of AF-related stroke by approximately two-thirds and can provide a 10% reduction in overall mortality. There has been increased interest in detecting AF due to its increased incidence and the possibility to prevent AF-related strokes. Inexpensive consumer devices which measure the ECG may have the potential to accurately detect AF but do not generally incorporate diagnostic algorithms. Machine learning algorithms have the potential to improve patient outcomes particularly where diagnoses are made from large volumes or complex patterns of data such as in AF. METHODS:We designed a novel AF detection algorithm using a de-correlated Lorenz plot of 60 consecutive RR intervals. In order to reduce the volume of data, the resulting images were compressed using a wavelet transformation (JPEG200 algorithm) and the compressed images were used as input data to a Support Vector Machine (SVM) classifier. We used the Massachusetts Institute of Technology (MIT)-Beth Israel Hospital (BIH) Atrial Fibrillation database and the MIT-BIH Arrhythmia database as training data and verified the algorithm performance using RR intervals collected using an inexpensive consumer heart rate monitor device (Polar-H7) in a case-control study. RESULTS:The SVM algorithm yielded excellent discrimination in the training data with a sensitivity of 99.2% and a specificity of 99.5% for AF. In the validation data, the SVM algorithm correctly identified AF in 79/79 cases; sensitivity 100% (95% CI 95.4%-100%) and non-AF in 328/336 cases; specificity 97.6% (95% CI 95.4%-99.0%). CONCLUSIONS:An inexpensive wearable heart rate monitor and machine learning algorithm can be used to detect AF with very high accuracy and has the capability to transmit ECG data which could be used to confirm AF. It could potentially be used for intermittent screening or continuously for prolonged periods to detect paroxysmal AF. Further work could lead to cost-effective and accurate estimation of AF burden and improved risk stratification in AF.
| S-EPMC6980577 | biostudies-literature
Project description:Atrial fibrillation (AF) is a common cardiac arrhythmia that affects 1% of the population worldwide and is associated with high levels of morbidity and mortality. Catheter ablation (CA) has become one of the first line treatments for AF, but its success rates are suboptimal, especially in the case of persistent AF. Computational approaches have shown promise in predicting the CA strategy using simulations of atrial models, as well as applying deep learning to atrial images. We propose a novel approach that combines image-based computational modelling of the atria with deep learning classifiers trained on patient-specific atrial models, which can be used to assist in CA therapy selection. Therefore, we trained a deep convolutional neural network (CNN) using a combination of (i) 122 atrial tissue images obtained by unfolding patient LGE-MRI datasets, (ii) 157 additional synthetic images derived from the patient data to enhance the training dataset, and (iii) the outcomes of 558 CA simulations to terminate several AF scenarios in the corresponding image-based atrial models. Four CNN classifiers were trained on this patient-specific dataset balanced using several techniques to predict three common CA strategies from the patient atrial images: pulmonary vein isolation (PVI), rotor-based ablation (Rotor) and fibrosis-based ablation (Fibro). The training accuracy for these classifiers ranged from 96.22 to 97.69%, while the validation accuracy was from 78.68 to 86.50%. After training, the classifiers were applied to predict CA strategies for an unseen holdout test set of atrial images, and the results were compared to outcomes of the respective image-based simulations. The highest success rate was observed in the correct prediction of the Rotor and Fibro strategies (100%), whereas the PVI class was predicted in 33.33% of the cases. In conclusion, this study provides a proof-of-concept that deep neural networks can learn from patient-specific MRI datasets and image-derived models of AF, providing a novel technology to assist in tailoring CA therapy to a patient.
| S-EPMC8187921 | biostudies-literature
Project description:Background: Radiofrequency catheter ablation (RFCA) therapy is the first-line treatment for atrial fibrillation (AF), the most common type of cardiac arrhythmia globally. However, the procedure currently has low success rates in dealing with persistent AF, with a reoccurrence rate of ∼50% post-ablation. Therefore, deep learning (DL) has increasingly been applied to improve RFCA treatment for AF. However, for a clinician to trust the prediction of a DL model, its decision process needs to be interpretable and have biomedical relevance. Aim: This study explores interpretability in DL prediction of successful RFCA therapy for AF and evaluates if pro-arrhythmogenic regions in the left atrium (LA) were used in its decision process. Methods: AF and its termination by RFCA have been simulated in MRI-derived 2D LA tissue models with segmented fibrotic regions (n = 187). Three ablation strategies were applied for each LA model: pulmonary vein isolation (PVI), fibrosis-based ablation (FIBRO) and a rotor-based ablation (ROTOR). The DL model was trained to predict the success of each RFCA strategy for each LA model. Three feature attribution (FA) map methods were then used to investigate interpretability of the DL model: GradCAM, Occlusions and LIME. Results: The developed DL model had an AUC (area under the receiver operating characteristic curve) of 0.78 ± 0.04 for predicting the success of the PVI strategy, 0.92 ± 0.02 for FIBRO and 0.77 ± 0.02 for ROTOR. GradCAM had the highest percentage of informative regions in the FA maps (62% for FIBRO and 71% for ROTOR) that coincided with the successful RFCA lesions known from the 2D LA simulations, but unseen by the DL model. Moreover, GradCAM had the smallest coincidence of informative regions of the FA maps with non-arrhythmogenic regions (25% for FIBRO and 27% for ROTOR). Conclusion: The most informative regions of the FA maps coincided with pro-arrhythmogenic regions, suggesting that the DL model leveraged structural features of MRI images to identify such regions and make its prediction. In the future, this technique could provide a clinician with a trustworthy decision support tool.
| S-EPMC10043207 | biostudies-literature
Project description:BackgroundArtificial intelligence (AI) models applied to 12-lead ECG waveforms can predict atrial fibrillation (AF), a heritable and morbid arrhythmia. However, the factors forming the basis of risk predictions from AI models are usually not well understood. We hypothesized that there might be a genetic basis for an AI algorithm for predicting the 5-year risk of new-onset AF using 12-lead ECGs (ECG-AI)-based risk estimates.MethodsWe applied a validated ECG-AI model for predicting incident AF to ECGs from 39 986 UK Biobank participants without AF. We then performed a genome-wide association study (GWAS) of the predicted AF risk and compared it with an AF GWAS and a GWAS of risk estimates from a clinical variable model.ResultsIn the ECG-AI GWAS, we identified 3 signals (P<5×10-8) at established AF susceptibility loci marked by the sarcomeric gene TTN and sodium channel genes SCN5A and SCN10A. We also identified 2 novel loci near the genes VGLL2 and EXT1. In contrast, the clinical variable model prediction GWAS indicated a different genetic profile. In genetic correlation analysis, the prediction from the ECG-AI model was estimated to have a higher correlation with AF than that from the clinical variable model.ConclusionsPredicted AF risk from an ECG-AI model is influenced by genetic variation implicating sarcomeric, ion channel and body height pathways. ECG-AI models may identify individuals at risk for disease via specific biological pathways.
| S-EPMC10524395 | biostudies-literature
Project description:BackgroundAdvances in ablation for atrial fibrillation (AF) continue to be hindered by ambiguities in mapping, even between experts. We hypothesized that convolutional neural networks (CNN) may enable objective analysis of intracardiac activation in AF, which could be applied clinically if CNN classifications could also be explained.MethodsWe performed panoramic recording of bi-atrial electrical signals in AF. We used the Hilbert-transform to produce 175 000 image grids in 35 patients, labeled for rotational activation by experts who showed consistency but with variability (kappa [κ]=0.79). In each patient, ablation terminated AF. A CNN was developed and trained on 100 000 AF image grids, validated on 25 000 grids, then tested on a separate 50 000 grids.ResultsIn the separate test cohort (50 000 grids), CNN reproducibly classified AF image grids into those with/without rotational sites with 95.0% accuracy (CI, 94.8%-95.2%). This accuracy exceeded that of support vector machines, traditional linear discriminant, and k-nearest neighbor statistical analyses. To probe the CNN, we applied gradient-weighted class activation mapping which revealed that the decision logic closely mimicked rules used by experts (C statistic 0.96).ConclusionsCNNs improved the classification of intracardiac AF maps compared with other analyses and agreed with expert evaluation. Novel explainability analyses revealed that the CNN operated using a decision logic similar to rules used by experts, even though these rules were not provided in training. We thus describe a scaleable platform for robust comparisons of complex AF data from multiple systems, which may provide immediate clinical utility to guide ablation. Registration: URL: https://www.clinicaltrials.gov; Unique identifier: NCT02997254. Graphic Abstract: A graphic abstract is available for this article.
| S-EPMC7438307 | biostudies-literature
Project description:Automatic bird detection in ornithological analyses is limited by the accuracy of existing models, due to the lack of training data and the difficulties in extracting the fine-grained features required to distinguish bird species. Here we apply the domain randomization strategy to enhance the accuracy of the deep learning models in bird detection. Trained with virtual birds of sufficient variations in different environments, the model tends to focus on the fine-grained features of birds and achieves higher accuracies. Based on the 100 terabytes of 2-month continuous monitoring data of egrets, our results cover the findings using conventional manual observations, e.g., vertical stratification of egrets according to body size, and also open up opportunities of long-term bird surveys requiring intensive monitoring that is impractical using conventional methods, e.g., the weather influences on egrets, and the relationship of the migration schedules between the great egrets and little egrets.
| S-EPMC7803967 | biostudies-literature
Project description:ObjectivePhotoplethysmography (PPG) monitoring has been implemented in many portable and wearable devices we use daily for health and fitness tracking. Its simplicity and cost-effectiveness has enabled a variety of biomedical applications, such as continuous long-term monitoring of heart arrhythmias, fitness, and sleep tracking, and hydration monitoring. One major issue that can hinder PPG-based applications is movement artifacts, which can lead to false interpretations. In many implementations, noisy PPG signals are discarded. Misinterpreted or discarded PPG signals pose a problem in applications where the goal is to increase the yield of detecting physiological events, such as in the case of paroxysmal atrial fibrillation (AF)-a common episodic heart arrhythmia and a leading risk factor for stroke. In this work, we compared a traditional machine learning and deep learning approaches for PPG quality assessment in the presence of AF, in order to find the most robust method for PPG quality assessment.ApproachThe training data set was composed of 78 278 30 s long PPG recordings from 3764 patients using bedside patient monitors. Two different representations of PPG signals were employed-a time-series based (1D) one and an image-based (2D) one. Trained models were tested on an independent set of 2683 30 s PPG signals from 13 stroke patients.Main resultsResNet18 showed a higher performance (0.985 accuracy, 0.979 specificity, and 0.988 sensitivity) than SVM and other deep learning approaches. 2D-based models were generally more accurate than 1D-based models.Significance2D representation of PPG signal enhances the accuracy of PPG signal quality assessment.
| S-EPMC7198064 | biostudies-literature