A Telesurveillance System With Automatic Electrocardiogram Interpretation Based on Support Vector Machine and Rule-Based Processing

Ho, Te-Wei; Huang, Chen‐Wei; Lin, Ching-Miao; Lai, Feipei; Ding, Jian-Jiun; Ho, Yi‐Lwun; Hung, Chi‐Sheng

doi:10.2196/medinform.4397

Cited by 15 publications

(8 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With this assumption, there are 109,494 annotated beats in this database. This number is consistent with the results of other studies [17,18,19,20,21,22,23,24,25]. To be consistent with other studies of R-peaks detection, quantitative comparisons in terms of sensitivity (S), positive predictivity (P), and detection error (DER), as defined in Equations (1)–(3), are studied.…”

Section: Methodssupporting

confidence: 89%

Low Resource Complexity R-peak Detection Based on Triangle Template Matching and Moving Average Filter

Nguyen

Dinh

Bui

2019

Sensors

View full text Add to dashboard Cite

A novel R-peak detection algorithm suitable for wearable electrocardiogram (ECG) devices is proposed with four objectives: robustness to noise, low latency processing, low resource complexity, and automatic tuning of parameters. The approach is a two-pronged algorithm comprising (1) triangle template matching to accentuate the slope information of the R-peaks and (2) a single moving average filter to define a dynamic threshold for peak detection. The proposed algorithm was validated on eight ECG public databases. The obtained results not only presented good accuracy, but also low resource complexity, all of which show great potential for detection R-peaks in ECG signals collected from wearable devices.

show abstract

Section: Methodssupporting

confidence: 89%

Low Resource Complexity R-peak Detection Based on Triangle Template Matching and Moving Average Filter

Nguyen

Dinh

Bui

2019

Sensors

View full text Add to dashboard Cite

show abstract

“…Large-scale machine learning methods have been investigated to reduce the human efforts for ECG beat classification [8][9][10]. However, most machine learning approaches with static and handcrafted features have performed at lower accuracy rates over new types of ECGs because those features are insufficient for representing the great diversity of ECG patterns from various patients.…”

Section: Introductionmentioning

confidence: 99%

A Lightweight Deep Learning Model for Fast Electrocardiographic Beats Classification With a Wearable Cardiac Monitor: Development and Validation Study

Jeon

Kwon

et al. 2020

JMIR Med Inform

View full text Add to dashboard Cite

Background Electrocardiographic (ECG) monitors have been widely used for diagnosing cardiac arrhythmias for decades. However, accurate analysis of ECG signals is difficult and time-consuming work because large amounts of beats need to be inspected. In order to enhance ECG beat classification, machine learning and deep learning methods have been studied. However, existing studies have limitations in model rigidity, model complexity, and inference speed. Objective To classify ECG beats effectively and efficiently, we propose a baseline model with recurrent neural networks (RNNs). Furthermore, we also propose a lightweight model with fused RNN for speeding up the prediction time on central processing units (CPUs). Methods We used 48 ECGs from the MIT-BIH (Massachusetts Institute of Technology-Beth Israel Hospital) Arrhythmia Database, and 76 ECGs were collected with S-Patch devices developed by Samsung SDS. We developed both baseline and lightweight models on the MXNet framework. We trained both models on graphics processing units and measured both models’ inference times on CPUs. Results Our models achieved overall beat classification accuracies of 99.72% for the baseline model with RNN and 99.80% for the lightweight model with fused RNN. Moreover, our lightweight model reduced the inference time on CPUs without any loss of accuracy. The inference time for the lightweight model for 24-hour ECGs was 3 minutes, which is 5 times faster than the baseline model. Conclusions Both our baseline and lightweight models achieved cardiologist-level accuracies. Furthermore, our lightweight model is competitive on CPU-based wearable hardware.

show abstract

“…An independent cardiologist performed the physician-based ECG interpretations. The computer-based ECG auto-interpretation was executed automatically according to the cloud-computing algorithm developed by the TELEHEALTH study group [ 29 ]. The institutional review board at NTUH approved the study protocol.…”

Section: Methodsmentioning

confidence: 99%

“…The telesurveillance system ( Figure 3 ) is a Web-based platform developed by the Graduate Institute of Biomedical Electronics and Bioinformatics, National Taiwan University, Taiwan [ 29 ]. The telesurveillance system was operated under a service-oriented architecture framework with the Health Level Seven standard.…”

Section: Methodsmentioning

confidence: 99%

“…The design and algorithm of the Web-based ECG auto-interpretation were described in a previous study [ 29 ]. After removing baseline noise by finite impulse response filter, the key features of the ECG waveforms extracted were processed by support vector machine or rule-based processing to construct a classification model that can suggest diagnosis.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Atrial Fibrillation Screening in Nonmetropolitan Areas Using a Telehealth Surveillance System With an Embedded Cloud-Computing Algorithm: Prospective Pilot Study

Chen

Hung

Huang

et al. 2017

JMIR Mhealth Uhealth

Self Cite

View full text Add to dashboard Cite

BackgroundAtrial fibrillation (AF) is a common form of arrhythmia that is associated with increased risk of stroke and mortality. Detecting AF before the first complication occurs is a recognized priority. No previous studies have examined the feasibility of undertaking AF screening using a telehealth surveillance system with an embedded cloud-computing algorithm; we address this issue in this study.ObjectiveThe objective of this study was to evaluate the feasibility of AF screening in nonmetropolitan areas using a telehealth surveillance system with an embedded cloud-computing algorithm.MethodsWe conducted a prospective AF screening study in a nonmetropolitan area using a single-lead electrocardiogram (ECG) recorder. All ECG measurements were reviewed on the telehealth surveillance system and interpreted by the cloud-computing algorithm and a cardiologist. The process of AF screening was evaluated with a satisfaction questionnaire.ResultsBetween March 11, 2016 and August 31, 2016, 967 ECGs were recorded from 922 residents in nonmetropolitan areas. A total of 22 (2.4%, 22/922) residents with AF were identified by the physician’s ECG interpretation, and only 0.2% (2/967) of ECGs contained significant artifacts. The novel cloud-computing algorithm for AF detection had a sensitivity of 95.5% (95% CI 77.2%-99.9%) and specificity of 97.7% (95% CI 96.5%-98.5%). The overall satisfaction score for the process of AF screening was 92.1%.ConclusionsAF screening in nonmetropolitan areas using a telehealth surveillance system with an embedded cloud-computing algorithm is feasible.

show abstract

A Telesurveillance System With Automatic Electrocardiogram Interpretation Based on Support Vector Machine and Rule-Based Processing

Cited by 15 publications

References 52 publications

Low Resource Complexity R-peak Detection Based on Triangle Template Matching and Moving Average Filter

Low Resource Complexity R-peak Detection Based on Triangle Template Matching and Moving Average Filter

A Lightweight Deep Learning Model for Fast Electrocardiographic Beats Classification With a Wearable Cardiac Monitor: Development and Validation Study

Atrial Fibrillation Screening in Nonmetropolitan Areas Using a Telehealth Surveillance System With an Embedded Cloud-Computing Algorithm: Prospective Pilot Study

Contact Info

Product

Resources

About