Densely Connected Convolutional Networks and Signal Quality Analysis to Detect Atrial Fibrillation Using Short Single-Lead ECG Recordings

Rubin, Jonathan; Parvaneh, Saman; Rahman, Asif; Conroy, Bryan; Babaeizadeh, Saeed

doi:10.22489/cinc.2017.160-246

Cited by 25 publications

(5 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This approach obtained of accuracy. In [11] signal quality index SQI technique was combined with CNN followed by a post-processing feature-based approach to classify AF. The accuracy for the PhysioNet/CinC database was .…”

Section: Detección De Fibrilación Auricular En Señales Ecg Usando Redes Neuronales Para Pacientes Específicosmentioning

confidence: 99%

See 1 more Smart Citation

Patient-Specific Detection of Atrial Fibrillation in Segments of ECG Signals using Deep Neural Networks

Castillo

Granados

Ariza

2019

Cien.Ing.Neogranadina

View full text Add to dashboard Cite

Atrial Fibrillation (AF) is the most common cardiac arrhythmia worldwide. It is associated with reduced quality of life and increases the risk of stroke and myocardial infarction. Unfortunately, many cases of AF are asymptomatic and undiagnosed, which increases the risk for the patients. Due to its paroxysmal nature, the detection of AF requires the evaluation, by a cardiologist, of long-term ECG signals. In Colombia, it is difficult to have access to an early diagnosis of AF because of the associated costs to the detection and the geographical distribution of cardiologists. This work is part of a macro project that aims to develop a specific-patient portable device for the detection of AF. This device will be based on a Convolutional Neural Network (CNN). We aim to find a suitable CNN model, which later could be implemented in hardware. Diverse techniques were applied to improve the answer regarding accuracy, sensitivity, specificity, and precision. The final model achieves an accuracy of , a specificity of , a sensitivity of and a precision of . During the development of the model, the computational cost and memory resources were taking into account in order to obtain an efficient hardware model in a future implementation of the device.

show abstract

Section: Detección De Fibrilación Auricular En Señales Ecg Usando Redes Neuronales Para Pacientes Específicosmentioning

confidence: 99%

“…As mentioned before, several computational CNN architectures have been developed [3], [10], [11], [20]; they focus on achieving higher accuracy. However, these works do not take into account issues regarding hardware and power consumption.…”

Section: Selecting the Cnn Architecture: A Hardware Point Of Viewmentioning

confidence: 99%

Patient-Specific Detection of Atrial Fibrillation in Segments of ECG Signals using Deep Neural Networks

Castillo

Granados

Ariza

2019

Cien.Ing.Neogranadina

View full text Add to dashboard Cite

show abstract

“…They concluded that their neural networks outperform their featurebased classifiers, showing the strength of the purely neural network-based approach. Parvaneh et al [18] improved a dense convolutional network by signal quality index and by the transformation of signal to the frequency domain. Their approach was similar to ours as they applied a neural network to extract frequency-domain features.…”

Section: Related Workmentioning

confidence: 99%

Strong deep learning baseline for single lead ECG processing

Csaba¹,

Hakkel²,

Goda³

et al. 2019

StuCoSReC. Proceedings of the 2019 6th Student Computer Science Research Conference

View full text Add to dashboard Cite

Objective: Atrial fibrillation (AF) is one of the most common serious abnormal heart rhythm conditions, and the number of deaths related to atrial fibrillation has increased by an order of magnitude in the past decades. We aim to create a system, which can provide help for cardiologist, classifying and highlighting important segments in recordings. Approach: In this paper, we propose a novel approach for AF detection using only a deep neural architecture without any traditional feature extractor for real-time automated suggestions of possible cardiac failures that can detect class invariant anomalies in signals recorded by a single channel portable ECG device. Results: Detecting the four categories: Normal, AF, Other and Noisy in terms of the official, F1 metric of hidden dataset maintained by the organizers of PhysioNet Computing in Cardiology Challenge 2017, our proposed algorithm has scored 0.88, 0.80, 0.69, 0.64 points respectively, and 0.79 on average.

show abstract

“…The availability of these large datasets allows for efficient training of DL algorithms. Since the seminal paper of Hannun et al (2019) several DL techniques have been suggested for the analysis of ECG signals, using convolutional neural networks (CNN) (Rubin et al 2017), recurrent neural networks (Teijeiro et al 2017) or more recently Transformer models (Vaswani et al 2017, Natarajan et al 2020). One of the current problems with DL approaches lies in the fact that such techniques do not easily generalize across new databases, where the behaviour of such automated classifiers is often unpredictable (Park et al 2021).…”

Section: Introductionmentioning

confidence: 99%

Predict alone, decide together: cardiac abnormality detection based on single lead classifier voting

et al. 2022

View full text Add to dashboard Cite

Objective: A classifier based on weighted voting of multiple single-lead based models combining deep learning representation and hand-crafted features was developed to classify 26 cardiac abnormalities from different lead subsets of short-term electrocardiograms (ECG). Approach: A two-stage method was proposed for the multilead prediction. First a lead-agnostic hybrid classifier was trained to predict the pathologies from single-lead ECG signals. The classifier combined fully automated deep learning (DL) features extracted through a convolutional neural network with hand-crafted features through a fully connected layer. Second, a voting of the single-lead based predictions was performed. For the 12-lead subset, voting consisted in an optimised weighting of the output probabilities of all available single lead predictions. For other lead subsets, voting simply consisted in the average of the lead predictions. Main Results: This approach achieved a challenge test score of 0.48, 0.47, 0.46, 0.46, 0.45 on the 12, 6, 4, 3, 2-lead subsets respectively on the 2021 Physionet/Computing in Cardiology challenge hidden test set. The use of an hybrid approach and more advanced voting layer improved some individual class classification but did not offer better generalization than our baseline fully DL approach. Significance: The proposed approach showed potential at correctly classifying main cardiac abnormalities and dealt well with reduced lead subsets.

show abstract

Densely Connected Convolutional Networks and Signal Quality Analysis to Detect Atrial Fibrillation Using Short Single-Lead ECG Recordings

Cited by 25 publications

References 15 publications

Patient-Specific Detection of Atrial Fibrillation in Segments of ECG Signals using Deep Neural Networks

Patient-Specific Detection of Atrial Fibrillation in Segments of ECG Signals using Deep Neural Networks

Strong deep learning baseline for single lead ECG processing

Predict alone, decide together: cardiac abnormality detection based on single lead classifier voting

Contact Info

Product

Resources

About