scholarly journals Machine learning detection of Atrial Fibrillation using wearable technology

PLoS ONE ◽  
2020 ◽  
Vol 15 (1) ◽  
pp. e0227401
Author(s):  
Mark Lown ◽  
Michael Brown ◽  
Chloë Brown ◽  
Arthur M. Yue ◽  
Benoy N. Shah ◽  
...  
Keyword(s):  
Machine Learning ◽  
Atrial Fibrillation ◽  
Wearable Technology

scholarly journals USING MACHINE LEARNING TO PREDICT READMISSIONS FOR PATIENTS WITH ATRIAL FIBRILLATION AFTER CATHETER ABLATION

2021 ◽  
Vol 77 (18) ◽  
pp. 331
Author(s):  
Anvi Raina ◽  
Shilpkumar Arora ◽  
Christopher DeSimone ◽  
Siva Mulpuru ◽  
Abhishek Deshmukh
Keyword(s):  
Machine Learning ◽  
Atrial Fibrillation ◽  
Catheter Ablation

scholarly journals Feasibility and accuracy of the automated software for dynamic quantification of left ventricular and atrial volumes and function in a large unselected population

2021 ◽  
Vol 22 (Supplement_1) ◽  
Author(s):  
G Italiano ◽  
G Tamborini ◽  
V Mantegazza ◽  
V Volpato ◽  
L Fusini ◽  
...  
Keyword(s):  
Machine Learning ◽  
Atrial Fibrillation ◽  
Image Quality ◽  
Normal Subjects ◽  
Left Ventricular ◽  
Systolic Volume ◽  
Unselected Population ◽  
Lv Volumes ◽  
Time Required ◽  
2D Images

Abstract Funding Acknowledgements Type of funding sources: None. Objective. Preliminary studies showed the accuracy of machine learning based automated dynamic quantification of left ventricular (LV) and left atrial (LA) volumes. We aimed to evaluate the feasibility and accuracy of machine learning based automated dynamic quantification of LV and LA volumes in an unselected population. Methods. We enrolled 600 unselected patients (12% in atrial fibrillation) clinically referred for transthoracic echocardiography (2DTTE), who also underwent 3D echocardiography (3DE) imaging. LV ejection fraction (EF), LV and LA volumes were obtained from 2D images; 3D images were analysed using Dynamic Heart Model (DHM) software (Philips) resulting in LV and LA volume-time curves. A subgroup of 140 patients underwent also cardiac magnetic resonance (CMR) imaging. Average time of analysis, feasibility, and image quality were recorded and results were compared between 2DTTE, DHM and CMR. Results. The use of DHM was feasible in 522/600 cases (87%). When feasible, the boundary position was considered accurate in 335/522 patients (64%), while major (n = 38) or minor (n = 149) borders corrections were needed. The overall time required for DHM datasets was approximately 40 seconds, resulting in physiologically appearing LV and LA volume–time curves in all cases. As expected, DHM LV volumes were larger than 2D ones (end-diastolic volume: 173 ± 64 vs 142 ± 58 mL, respectively), while no differences were found for LV EF and LA volumes (EF: 55%±12 vs 56%±14; LA volume 89 ± 36 vs 89 ± 38 mL, respectively). The comparison between DHM and CMR values showed a high correlation for LV volumes (r = 0.70 and r = 0.82, p < 0.001 for end-diastolic and end-systolic volume, respectively) and an excellent correlation for EF (r= 0.82, p < 0.001) and LA volumes. Conclusions. The DHM software is feasible, accurate and quick in a large series of unselected patients, including those with suboptimal 2D images or in atrial fibrillation. Table 1 DHM quality Adjustment Feasibility Good Suboptimal Minor Major Total of patients (n, %) 522/600 (87%) 327/522 (62%) 195/522 (28%) 149/522 (29%) 38/522 (6%) Normal subjects (n, %) 39/40 (97%) 23/39 (57%) 16/39 (40%) 9/39 (21%) 1/39 (3%) Atrial Fibrillation (n, %) 59/73 (81%)* 28/59 (47%) 31/59 (53%) 15/59 (25%) 6/59 (10%) Valvular disease (n, %) 271/312 (87%) 120/271 (%) 151/271 (%) 65/271 (24%) 16/271 (6%) Coronary artery disease (n, %) 47/58 (81%)* 26/47 (46%) 21/47 (37%) 16/47 (34%) 5/47 (11%) Miscellaneous (n, %) 24/25 (96%) 18/24 (75%) 6/24 (25%) 5/24 (21%) 3/24 (12%) Feasibility of DHM, image quality and need to adjustments in global population and in each subgroup. Abstract Figure 1

scholarly journals Prediction model for thyrotoxic atrial fibrillation: a retrospective study.

2021 ◽  
Author(s):  
Daria Aleksandrovna Ponomartseva ◽  
Ilia Vladislavovich Derevitskii ◽  
Sergey Valerevich Kovalchuk ◽  
Alina Yurevna Babenko
Keyword(s):  
Machine Learning ◽  
Atrial Fibrillation ◽  
Heart Rate ◽  
Retrospective Study ◽  
Prediction Model ◽  
Performance Metrics ◽  
Premature Ventricular Contraction ◽  
Ventricular Contraction ◽  
Atrial Contraction ◽  
Premature Atrial Contraction

Abstract Background: Thyrotoxic atrial fibrillation (TAF) is a recognized significant complication of hyperthyroidism. Early identification of the individuals predisposed to TAF would improve thyrotoxic patients’ management. However, to our knowledge, an instrument that establishes an individual risk of the condition is unavailable. Therefore, the aim of this study is to build a TAF prediction model and rank TAF predictors in order of importance. Methods: In this retrospective study, we have investigated 36 demographic and clinical features for 420 patients with overt hyperthyroidism, 30% of which had TAF. At first, the association of these features with TAF was evaluated by classical statistical methods. Then, we developed several TAF prediction models with eight different machine learning classifiers and compared them by performance metrics. The models included ten features that were selected based on their clinical effectuality and importance for model output. Finally, we ranked TAF predictors, elicited from the optimal final model, by the machine learning tehniques. Results: The best performance metrics prediction model was built with the extreme gradient boosting classifier. It had the reasonable accuracy of 84% and AUROC of 0.89 on the test set. The model confirmed such well-known TAF risk factors as age, sex, hyperthyroidism duration, heart rate and some concomitant cardiovascular diseases (arterial hypertension and conjestive heart rate). We also identified premature atrial contraction and premature ventricular contraction as new TAF predictors. The top five TAF predictors, elicited from the model, included (in order of importance) PAC, PVC, hyperthyroidism duration, heart rate during hyperthyroidism and age. Conclusions: We developed a machine learning model for TAF prediction. It seems to be the first available analytical tool for TAF risk assessment. In addition, we defined five most important TAF predictors, including premature atrial contraction and premature ventricular contraction as the new ones. These results have contributed to TAF prediction investigation and may serve as a basis for further research focused on TAF prediction improvement and facilitation of thyrotoxic patients’ management.

scholarly journals Interpretable Machine Learning for Early Neurological Deterioration Prediction in Atrial Fibrillation-Related Stroke

2021 ◽  
Author(s):  
Seong Hwan Kim ◽  
Eun-Tae Jeon ◽  
Sungwook Yu ◽  
Kyungmi O ◽  
Chi Kyung Kim ◽  
...  
Keyword(s):  
Machine Learning ◽  
Atrial Fibrillation ◽  
Neurological Deterioration ◽  
Gradient Boosting ◽  
Support Vector ◽  
Light Gradient ◽  
Interpretable Machine Learning ◽  
Extreme Gradient Boosting ◽  
Early Neurological Deterioration ◽  
Feature Importance

Abstract We aimed to develop a novel prediction model for early neurological deterioration (END) based on an interpretable machine learning (ML) algorithm for atrial fibrillation (AF)-related stroke and to evaluate the prediction accuracy and feature importance of ML models. Data from multi-center prospective stroke registries in South Korea were collected. After stepwise data preprocessing, we utilized logistic regression, support vector machine, extreme gradient boosting, light gradient boosting machine (LightGBM), and multilayer perceptron models. We used the Shapley additive explanations (SHAP) method to evaluate feature importance. Of the 3,623 stroke patients, the 2,363 who had arrived at the hospital within 24 hours of symptom onset and had available information regarding END were included. Of these, 318 (13.5%) had END. The LightGBM model showed the highest area under the receiver operating characteristic curve (0.778, 95% CI, 0.726 - 0.830). The feature importance analysis revealed that fasting glucose level and the National Institute of Health Stroke Scale score were the most influential factors. Among ML algorithms, the LightGBM model was particularly useful for predicting END, as it revealed new and diverse predictors. Additionally, the SHAP method can be adjusted to individualize the features’ effects on the predictive power of the model.

Diagnosis of Atrial Fibrillation Using Machine Learning with Wearable Devices after Cardiac Surgery (Preprint)

2021 ◽  
Author(s):  
Daisuke Hiraoka ◽  
Tomohiko Inui ◽  
Eiryo Kawakami ◽  
Megumi Oya ◽  
Ayumu Tsuji ◽  
...  
Keyword(s):  
Machine Learning ◽  
Atrial Fibrillation ◽  
Cardiac Surgery ◽  
Early Detection ◽  
Pulse Rate ◽  
Paroxysmal Atrial Fibrillation ◽  
Wearable Device ◽  
Strongly Correlated ◽  
Cross Correlation Analysis ◽  
Data Output

BACKGROUND Some attempts have been made to detect atrial fibrillation with a wearable device equipped with photoelectric volumetric pulse wave technology, and it is expected to be applied under real clinical conditions. OBJECTIVE This study is the second part of a two-phase study aimed at developing a method for immediate detection of paroxysmal atrial fibrillation (AF) using a wearable device with built-in PPG. The objective of this study is to develop an algorithm to immediately diagnose atrial fibrillation by wearing an Apple Watch equipped with a photoplethysmography (PPG) sensor on patients undergoing cardiac surgery and using machine learning of the pulse data output from the device. METHODS A total of 80 subjects who underwent cardiac surgery at a single institution between June 2020 and March 2021 were monitored for postoperative atrial fibrillation using telemetry monitored ECG and Apple Watch. Atrial fibrillation was diagnosed by qualified physicians from telemetry-monitored ECGs and 12-lead ECGs; a diagnostic algorithm was developed using machine learning on pulse rate data output from the Apple Watch. RESULTS One of the 80 patients was excluded from the analysis due to redness of the Apple Watch wearer. 27 (34.2%) of the 79 patients developed AF, and 199 events of AF, including brief AF, were observed. 18 events of AF lasting longer than 1 hour were observed, and Cross-correlation analysis (CCF) showed that pulse rate measured by Apple Watch was strongly correlated (CCF 0.6-0.8) with 8 events and very strongly correlated (CCF >0.8) with 3 events. The diagnostic accuracy by machine learning was 0.7952 (sensitivity 0.6312, specificity 0.8605 at the point closest to the top-left) for the AUC of the ROC curve. CONCLUSIONS We were able to safely monitor pulse rate in patients after cardiac surgery by wearing an Apple Watch. Although the pulse rate from the PPG sensor does not follow the heart rate of the telemetry monitoring ECG in some parts, which may reduce the accuracy of the diagnosis of atrial fibrillation by machine learning, we have shown the possibility of clinical application of early detection of atrial fibrillation using only the pulse rate collected by the PPG sensor. CLINICALTRIAL The use of wristband type continuous pulse measurement device with artificial intelligence for early detection of paroxysmal atrial fibrillation Clinical Research Protocol No. jRCTs032200032 https://jrct.niph.go.jp/latest-detail/jRCTs032200032

Sign in / Sign up

Export Citation Format

Share Document