Dense Convolutional Networks With Focal Loss and Image Generation for Electrocardiogram Classification

Rahhal, Mohamad Mahmoud Al; Bazi, Yakoub; Almubarak, Haidar; Alajlan, Naif

doi:10.1109/access.2019.2960116

Cited by 30 publications

(13 citation statements)

References 62 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Tran et al [42] describe a classifier of lung nodules using deep learning, reporting an accuracy improvement from 95.6% when using cross-entropy loss to 97.2% when using the focal loss. Similarly, the work published by Al et al [43] which evaluated the cross-entropy, over-sampling of the minority class, and the focal loss when using an imbalanced dataset demonstrated that focal loss produced better results.…”

Section: Class Imbalance Problemmentioning

confidence: 77%

“…As it can be seen in Table 2, the distribution of our dataset was skewed towards the negative class. According to the relevant literature, handling the class imbalance has been reported to improve the classification performance [41]- [43]. To verify this assertion, we conducted an experiment in This work is licensed under a Creative Commons Attribution 4.0 License.…”

Section: ) Handling Of the Class Imbalancementioning

confidence: 94%

See 1 more Smart Citation

Epiretinal Membrane Detection in Optical Coherence Tomography Retinal Images Using Deep Learning

et al. 2021

View full text Add to dashboard Cite

Epiretinal membrane (ERM) is an eye disease that affects 7% of the world population, with a higher incidence in people over 75 years old. If left untreated, it can lead to complications in the central vision, resulting in severe vision loss. Early detection is important for progress follow-up, treatment monitoring, and to avoid total vision loss. Optical coherence tomography, a non-invasive retina imaging technique, can be used for effective detection and monitoring of this condition. To date, automatic methods to detect ERM have received little attention in the research literature. This article describes the application of deep learning to the automatic detection of ERM. The proposed solution is based on four widely used convolutional neural network architectures adapted to the task using transfer learning, and fine-tuned with a proprietary dataset. The architectures were specialized by optimizing the network hyperparameters and two loss functions, cross-entropy and focal loss. A detailed description of the methods is provided, complemented with an exhaustive evaluation of their performance. Overall, the methods reached an accuracy of 99.7%, with sensitivity and specificity of 99.47% and 99.93%, respectively. The results showed that transfer learning enabled a successful use of deep learning to detect ERM in optical coherence tomography retinal images, even when only relatively small training datasets are available.INDEX TERMS Artificial intelligence, deep learning, epiretinal membrane, macular puker, neural networks, optical coherence tomography, transfer learning.

show abstract

Section: Class Imbalance Problemmentioning

confidence: 77%

Section: ) Handling Of the Class Imbalancementioning

confidence: 94%

Epiretinal Membrane Detection in Optical Coherence Tomography Retinal Images Using Deep Learning

et al. 2021

View full text Add to dashboard Cite

show abstract

“…In [38], Generative neural network is used to convert the raw 1D ECG signal data into a 2D image. These images are input to DenseNet which produces highly accurate classification, with high sensitivity and specificity using 4 classes of heart beat detection.…”

Section: B Two-dimensional Cnn Approachesmentioning

confidence: 99%

ECG Heartbeat Classification Using Multimodal Fusion

et al. 2021

View full text Add to dashboard Cite

Electrocardiogram (ECG) is an authoritative source to diagnose and counter critical cardiovascular syndromes such as arrhythmia and myocardial infarction (MI). Current machine learning techniques either depend on manually extracted features or large and complex deep learning networks which merely utilize the 1D ECG signal directly. Since intelligent multimodal fusion can perform at the state-of-the-art level with an efficient deep network, therefore, in this paper, we propose two computationally efficient multimodal fusion frameworks for ECG heart beat classification called Multimodal Image Fusion (MIF) and Multimodal Feature Fusion (MFF). At the input of these frameworks, we convert the raw ECG data into three different images using Gramian Angular Field (GAF), Recurrence Plot (RP) and Markov Transition Field (MTF). In MIF, we first perform image fusion by combining three imaging modalities to create a single image modality which serves as input to the Convolutional Neural Network (CNN). In MFF, we extracted features from penultimate layer of CNNs and fused them to get unique and interdependent information necessary for better performance of classifier. These informational features are finally used to train a Support Vector Machine (SVM) classifier for ECG heart-beat classification. We demonstrate the superiority of the proposed fusion models by performing experiments on PhysioNet's MIT-BIH dataset for five distinct conditions of arrhythmias which are consistent with the AAMI EC57 protocols and on PTB diagnostics dataset for Myocardial Infarction (MI) classification. We achieved classification accuracy of 99.7% and 99.2% on arrhythmia and MI classification, respectively. Source code at https://github.com/zaamad/ECG-Heartbeat-Classification-Using-Multimodal-Fusion INDEX TERMS Convolutional neural network, deep learning, ECG, image fusion, multimodal fusion.

show abstract

“…Deep learning techniques have recently proven their superiority over traditional approaches in image classification problems [14,15]. Deep learning techniques have also proven their advantages on 1D signals, including ECG [16][17][18][19]. Several studies have investigated the utilization of deep learning techniques in biometrics systems [20][21][22], and for fingerprint PAD [23][24][25][26].…”

Section: Introductionmentioning

confidence: 99%

End-to-End Deep Learning Fusion of Fingerprint and Electrocardiogram Signals for Presentation Attack Detection

Jomaa

Mathkour

Bazi

et al. 2020

Sensors

Self Cite

View full text Add to dashboard Cite

Although fingerprint-based systems are the commonly used biometric systems, they suffer from a critical vulnerability to a presentation attack (PA). Therefore, several approaches based on a fingerprint biometrics have been developed to increase the robustness against a PA. We propose an alternative approach based on the combination of fingerprint and electrocardiogram (ECG) signals. An ECG signal has advantageous characteristics that prevent the replication. Combining a fingerprint with an ECG signal is a potentially interesting solution to reduce the impact of PAs in biometric systems. We also propose a novel end-to-end deep learning-based fusion neural architecture between a fingerprint and an ECG signal to improve PA detection in fingerprint biometrics. Our model uses state-of-the-art EfficientNets for generating a fingerprint feature representation. For the ECG, we investigate three different architectures based on fully-connected layers (FC), a 1D-convolutional neural network (1D-CNN), and a 2D-convolutional neural network (2D-CNN). The 2D-CNN converts the ECG signals into an image and uses inverted Mobilenet-v2 layers for feature generation. We evaluated the method on a multimodal dataset, that is, a customized fusion of the LivDet 2015 fingerprint dataset and ECG data from real subjects. Experimental results reveal that this architecture yields a better average classification accuracy compared to a single fingerprint modality.

show abstract

Dense Convolutional Networks With Focal Loss and Image Generation for Electrocardiogram Classification

Cited by 30 publications

References 62 publications

Epiretinal Membrane Detection in Optical Coherence Tomography Retinal Images Using Deep Learning

Epiretinal Membrane Detection in Optical Coherence Tomography Retinal Images Using Deep Learning

ECG Heartbeat Classification Using Multimodal Fusion

End-to-End Deep Learning Fusion of Fingerprint and Electrocardiogram Signals for Presentation Attack Detection

Contact Info

Product

Resources

About