Scalable customization of atrial fibrillation detection in cardiac monitoring devices: Increasing detection accuracy through personalized monitoring in large patient populations

2011 ◽  
Author(s):  
Kuk Jin Jang ◽  
G. Balakrishnan ◽  
Z. Syed ◽  
N. Verma
Keyword(s):  
Atrial Fibrillation ◽  
Cardiac Monitoring ◽  
Detection Accuracy ◽  
Large Patient ◽  
Monitoring Devices ◽  
Atrial Fibrillation Detection

First clinical experience with the wearable cardiac rhythm long-term monitoring cardioskin device

2020 ◽  
Vol 41 (Supplement_2) ◽  
Author(s):  
P Guedeney ◽  
J Silvain ◽  
F Hidden-Lucet ◽  
C Maupain ◽  
S Dinanian ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Clinical Practice ◽  
Cardiac Rhythm ◽  
Cardiac Monitoring ◽  
Funding Source ◽  
Single Center ◽  
Long Term Monitoring ◽  
Monitoring Devices ◽  
Term Monitoring

Abstract Background There are only limited options for long-term cardiac monitoring devices readily available in clinical practice for outpatients. Holter monitoring devices are limited by the uncomfort of wires and patches, the small number of leads for analysis, the quality of recordings or the monitoring duration while insertable cardiac monitors are costly and exposed to potential local complication. Purpose To describe a single center experience with a novel wearable device for cardiac rhythm monitoring. Methods The Cardioskin™ system is a patch-free, wire-free, wearable device with rechargeable batteries that provides a high quality 15-lead electrocardiogram monitoring over 1 month (Figure 1). Data are sent using a mobile application downloaded in the patient smartphone to a central Corelab where they can be interpreted by an expert and/or the prescribing physician. An alarm signal is readily available within the Cardioskin™ device, to allow patients to indicate the presence of symptoms. In this single center retrospective registry, we provide a first report of the use of this novel device in real world practice, with indication and duration of cardiac monitoring left at the physicans “discretion”. Results From January 2019 to December 2019, the Cardioskin™ system was prescribed in 60 patients for an overall median duration of 26.5 (14–32) days. The mean age of the patients was 45±12.2 years and 24 (40%) were male. Indications for cardiac monitoring were post-Stroke, palpitation, syncope and cardiomyopathy assessment in 56%, 30%, 7% and 7% of the cases, respectively. A sustained (>30 seconds) supraventricular tachycardia was detected in 4 cases, including one case of atrial fibrillation, two case of atrial tachycardia and on case of junctional tachycardia. Unsustained ventricular tachycardia and atrial fibrillation burst were detected in another 2 cases (Figure 1). There was no reported case of skin irritation by the Cardioskin™ system or abrupt interruption of the monitoring by the patients. Conclusion The Cardioskin™ system is a novel, discreet and comfortable cardiac rhythm wearable long-term monitoring device which can be used in clinical practice for broad diagnostic indications. Figure 1. Cardioskin system Funding Acknowledgement Type of funding source: Public Institution(s). Main funding source(s): ACTION coeur

On-device Prior Knowledge Incorporated Learning for Personalized Atrial Fibrillation Detection

2021 ◽  
Vol 20 (5s) ◽  
pp. 1-25
Author(s):  
Zhenge Jia ◽  
Yiyu Shi ◽  
Samir Saba ◽  
Jingtong Hu
Keyword(s):  
Atrial Fibrillation ◽  
Prior Knowledge ◽  
Domain Knowledge ◽  
Fine Tuning ◽  
Cardiac Monitoring ◽  
Patient Specific ◽  
Detection Accuracy ◽  
Specific Patient ◽  
Deep Model ◽  
Tuning Strategy

Atrial Fibrillation (AF), one of the most prevalent arrhythmias, is an irregular heart-rate rhythm causing serious health problems such as stroke and heart failure. Deep learning based methods have been exploited to provide an end-to-end AF detection by automatically extracting features from Electrocardiogram (ECG) signal and achieve state-of-the-art results. However, the pre-trained models cannot adapt to each patient’s rhythm due to the high variability of rhythm characteristics among different patients. Furthermore, the deep models are prone to overfitting when fine-tuned on the limited ECG of the specific patient for personalization. In this work, we propose a prior knowledge incorporated learning method to effectively personalize the model for patient-specific AF detection and alleviate the overfitting problems. To be more specific, a prior-incorporated portion importance mechanism is proposed to enforce the network to learn to focus on the targeted portion of the ECG, following the cardiologists’ domain knowledge in recognizing AF. A prior-incorporated regularization mechanism is further devised to alleviate model overfitting during personalization by regularizing the fine-tuning process with feature priors on typical AF rhythms of the general population. The proposed personalization method embeds the well-defined prior knowledge in diagnosing AF rhythm into the personalization procedure, which improves the personalized deep model and eliminates the workload of manually adjusting parameters in conventional AF detection method. The prior knowledge incorporated personalization is feasibly and semi-automatically conducted on the edge, device of the cardiac monitoring system. We report an average AF detection accuracy of 95.3% of three deep models over patients, surpassing the pre-trained model by a large margin of 11.5% and the fine-tuning strategy by 8.6%.

scholarly journals Ultra‐early continuous cardiac monitoring improves atrial fibrillation detection and prognosis of patients with cryptogenic stroke

2019 ◽  
Vol 27 (2) ◽  
pp. 244-250 ◽  
Author(s):  
E. Cuadrado‐Godia ◽  
B. Benito ◽  
A. Ois ◽  
E. Vallès ◽  
A. Rodríguez‐Campello ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Cryptogenic Stroke ◽  
Cardiac Monitoring ◽  
Atrial Fibrillation Detection

Electromechanical cardiac monitoring SoC for atrial fibrillation detection

2017 ◽  
Author(s):  
Jonas Eriksson ◽  
Mika Kutila ◽  
Tapani Nevalainen ◽  
Phong Nguyen ◽  
Kati Sairanen ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Cardiac Monitoring ◽  
Atrial Fibrillation Detection

scholarly journals Wrist Band Photoplethysmography Autocorrelation Analysis Enables Detection of Atrial Fibrillation Without Pulse Detection

2021 ◽  
Vol 12 ◽  
Author(s):  
Eemu-Samuli Väliaho ◽  
Pekka Kuoppa ◽  
Jukka A. Lipponen ◽  
Juha E. K. Hartikainen ◽  
Helena Jäntti ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Characteristic Curve ◽  
Pulse Interval ◽  
Detection Accuracy ◽  
Monitoring Method ◽  
Pulse Detection ◽  
Detection Algorithms ◽  
Pulse Waves ◽  
Predictor Model ◽  
Atrial Fibrillation Detection

Atrial fibrillation is often asymptomatic and intermittent making its detection challenging. A photoplethysmography (PPG) provides a promising option for atrial fibrillation detection. However, the shapes of pulse waves vary in atrial fibrillation decreasing pulse and atrial fibrillation detection accuracy. This study evaluated ten robust photoplethysmography features for detection of atrial fibrillation. The study was a national multi-center clinical study in Finland and the data were combined from two broader research projects (NCT03721601, URL: https://clinicaltrials.gov/ct2/show/NCT03721601 and NCT03753139, URL: https://clinicaltrials.gov/ct2/show/NCT03753139). A photoplethysmography signal was recorded with a wrist band. Five pulse interval variability, four amplitude features and a novel autocorrelation-based morphology feature were calculated and evaluated independently as predictors of atrial fibrillation. A multivariate predictor model including only the most significant features was established. The models were 10-fold cross-validated. 359 patients were included in the study (atrial fibrillation n = 169, sinus rhythm n = 190). The autocorrelation univariate predictor model detected atrial fibrillation with the highest area under receiver operating characteristic curve (AUC) value of 0.982 (sensitivity 95.1%, specificity 93.7%). Autocorrelation was also the most significant individual feature (p < 0.00001) in the multivariate predictor model, detecting atrial fibrillation with AUC of 0.993 (sensitivity 96.4%, specificity 96.3%). Our results demonstrated that the autocorrelation independently detects atrial fibrillation reliably without the need of pulse detection. Combining pulse wave morphology-based features such as autocorrelation with information from pulse-interval variability it is possible to detect atrial fibrillation with high accuracy with a commercial wrist band. Photoplethysmography wrist bands accompanied with atrial fibrillation detection algorithms utilizing autocorrelation could provide a computationally very effective and reliable wearable monitoring method in screening of atrial fibrillation.

scholarly journals Cardiac monitoring for detection of atrial fibrillation after TIA: A systematic review and meta-analysis

2016 ◽  
Vol 12 (1) ◽  
pp. 33-45 ◽  
Author(s):  
Eleni Korompoki ◽  
Angela Del Giudice ◽  
Steffi Hillmann ◽  
Uwe Malzahn ◽  
David J Gladstone ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Systematic Review ◽  
Detection Rate ◽  
Meta Analysis ◽  
Cardiac Monitoring ◽  
Significant Heterogeneity ◽  
Detection Rates ◽  
Ischemic Attack ◽  
Meta Analyses ◽  
Atrial Fibrillation Detection

Background and purpose The detection rate of atrial fibrillation has not been studied specifically in transient ischemic attack (TIA) patients although extrapolation from ischemic stroke may be inadequate. We conducted a systematic review and meta-analysis to determine the rate of newly diagnosed atrial fibrillation using different methods of ECG monitoring in TIA. Methods A comprehensive literature search was performed following a pre-specified protocol the PRISMA statement. Prospective observational studies and randomized controlled trials were considered that included TIA patients who underwent cardiac monitoring for >12 h. Primary outcome was frequency of detection of atrial fibrillation ≥30 s. Analyses of subgroups and of duration and type of monitoring were performed. Results Seventeen studies enrolling 1163 patients were included. The pooled atrial fibrillation detection rate for all methods was 4% (95% CI: 2–7%). Yield of monitoring was higher in selected (higher age, more extensive testing for arrhythmias before enrolment, or presumed cardioembolic/cryptogenic cause) than in unselected cohorts (7% vs 3%). Pooled mean atrial fibrillation detection rates rose with duration of monitoring: 4% (24 h), 5% (24 h to 7 days) and 6% (>7 days), respectively. Yield of non-invasive was significantly lower than that of invasive monitoring (4% vs. 11%). Significant heterogeneity was observed among studies (I2=60.61%). Conclusion This first meta-analysis of atrial fibrillation detection in TIA patients finds a lower atrial fibrillation detection rate in TIA than reported for IS and TIA cohorts in previous meta-analyses. Prospective studies are needed to determine actual prevalence of atrial fibrillation and optimal diagnostic procedure for atrial fibrillation detection in TIA.

Sign in / Sign up

Export Citation Format

Share Document