A wavelet-based capsule neural network for ECG biometric identification

Boujnouni, Imane El; Zili, Hassan; Tali, Abdelhak; Tali, Tarik; Laaziz, Yassin

doi:10.1016/j.bspc.2022.103692

Cited by 20 publications

(10 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This model is only for five classes, but our proposed model identifies 156 subjects in a combined dataset. In [ 42 ], a method was proposed for converting ECG signals into scalogram images using the continuous wavelet transform (CWT) and then extracting two-dimensional features using the discrete wavelet transform (DWT) and passing them to the CNN model. Ref.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Hybrid Deep Learning and Discrete Wavelet Transform-Based ECG Biometric Recognition for Arrhythmic Patients and Healthy Controls

Asif

Faisal

Dar

et al. 2023

Sensors

View full text Add to dashboard Cite

The intrinsic and liveness detection behavior of electrocardiogram (ECG) signals has made it an emerging biometric modality for the researcher with several applications including forensic, surveillance and security. The main challenge is the low recognition performance with datasets of large populations, including healthy and heart-disease patients, with a short interval of an ECG signal. This research proposes a novel method with the feature-level fusion of the discrete wavelet transform and a one-dimensional convolutional recurrent neural network (1D-CRNN). ECG signals were preprocessed by removing high-frequency powerline interference, followed by a low-pass filter with a cutoff frequency of 1.5 Hz for physiological noises and by baseline drift removal. The preprocessed signal is segmented with PQRST peaks, while the segmented signals are passed through Coiflets’ 5 Discrete Wavelet Transform for conventional feature extraction. The 1D-CRNN with two long short-term memory (LSTM) layers followed by three 1D convolutional layers was applied for deep learning-based feature extraction. These combinations of features result in biometric recognition accuracies of 80.64%, 98.81% and 99.62% for the ECG-ID, MIT-BIH and NSR-DB datasets, respectively. At the same time, 98.24% is achieved when combining all of these datasets. This research also compares conventional feature extraction, deep learning-based feature extraction and a combination of these for performance enhancement, compared to transfer learning approaches such as VGG-19, ResNet-152 and Inception-v3 with a small segment of ECG data.

show abstract

Section: Resultsmentioning

confidence: 99%

“…Experiments show that, using this method, the identification results for 21 subjects can reach 99.285%. In [ 42 ], techniques are proposed for recognizing humans using ECG based on a combination of discrete wavelet transform, continuous wavelet transform and a novel deep learning approach called the capsule network.…”

Section: Related Workmentioning

confidence: 99%

Hybrid Deep Learning and Discrete Wavelet Transform-Based ECG Biometric Recognition for Arrhythmic Patients and Healthy Controls

Asif

Faisal

Dar

et al. 2023

Sensors

View full text Add to dashboard Cite

show abstract

“…Since then, ECG identification has gained significant attention and has been widely studied. Boujnouni et al [15] employed a capsule neural network (CapsNet) for individual identification, achieving impressive results of 98.8% accuracy on the PTB database. Fatimah et al [16] employed Fourier decomposition (FDM) and phase transform (PT) techniques to process the ECG signal and extract relevant features.…”

Section: Related Workmentioning

confidence: 99%

ECG Biometrics Based on Attention Enhanced Domain Adaptive Feature Fusion Network

Yi,

Si,

Fan

et al. 2024

IEEE Access

View full text Add to dashboard Cite

In recent years, there has been an increasing focus on privacy and security among individuals. Biometric systems are highly regarded for their strong resistance to counterfeiting. Among various biometric features, ECG signals are difficult to falsify and are less attack-prone. However, as the time interval between signal acquisitions increases, the dissimilarities between individual ECG signals become more pronounced, making it difficult for many studies to achieve satisfactory recognition results in multi-session recognition. Furthermore, some studies encountered challenges in extracting crucial features from the ECG signal, which posed difficulties for identification experiments. To address the above challenge, this study proposes a novel attention-enhanced domain adaptive feature fusion network. Firstly, the network employs a multi-branch architecture to extract essential features from various dimensions of the ECG signal. Secondly, it incorporates the proposed weight fusion adaptive attention mechanism to further emphasize the features of heartbeats that contribute to recognition. Additionally, domain adaptive technology is employed to mitigate differences in feature distribution of ECG signals across sessions, thereby enhancing the model's generalization capability. Finally, four well-known databases, the ECG-ID database, PTB database, CYBHi database, and Heartprint database, were utilized to evaluate the performance of the model. These databases predominantly comprise individuals with multiple records, fulfilling the prerequisites for multi-session recognition. The model achieved recognition results of 96.31%, 73.79%, 77.80%, and 54.78% on these four databases, respectively, surpassing related studies in multisession recognition scenarios.INDEX TERMS Electrocardiogram, biometrics, multi-branch architecture, weight fusion adaptive attention mechanism, domain adaptation.

show abstract

“…Recently, a new model of artificial neural networks, Capsule Neural Network (CapsNet), has been used in many classification problems. The CapsNet has been used for classification and detection in other areas, as in 28–32 . Hinton et al 33 proposed the capsule theory to solve some of CNN's disadvantages in 2011.…”

Section: Introductionmentioning

confidence: 99%

“…The CapsNet has been used for classification and detection in other areas, as in. [28][29][30][31][32] Hinton et al 33 proposed the capsule theory to solve some of CNN's disadvantages in 2011. The motivation for this study is the robust performance of the Caps-Net architecture on small, large, complex, and unbalanced data.…”

Section: Introductionmentioning

confidence: 99%

Examining the classification performance of pre‐trained capsule networks on imbalanced bone marrow cell dataset

Aydin Atasoy,

Faris Abdulla Al Rahhawi

2024

Int J Imaging Syst Tech

View full text Add to dashboard Cite

The automatic detection of bone marrow (BM) cell diseases plays a vital role in the medical field; it helps to make diagnoses more precise and effective, which leads to early detection and can significantly improve patient outcomes and increase the chances of successful intervention. This study proposed a fully automated intelligent system for BM classification by developing and enhancing Capsule Neural Network (CapsNet) architecture. Although CapsNet has demonstrated success in many classification tasks, it still has some limitations and challenges associated with using Convolutional Neural Networks (CNNs), which suffer from information loss during the pooling and discarding of detailed spatial information, resulting in the loss of fine‐grained features. Additionally, CNNs must help capture hierarchical feature relationships and often learn them implicitly by stacking convolutional layers. In contrast, CapsNets are designed to capture hierarchical features through dynamic routing and relationships between capsules, resulting in a more explicit representation of spatial hierarchy. CapsNets manage transformations and offer equivariance, preserving spatial information through capsule routing mechanisms. Further, to improve how features are represented, pre‐trained models such as Residual Capsule Network (RES‐CapsNet), Visual Geometry Group Capsule Network (VGG‐CapsNet), and Google Network (Inception V3) (GN‐CapsNet) have been used. This helps the network obtain the low‐ and mid‐level features and information it has previously learned so that subsequent capsule layers receive better initial information. Additionally, the Synthetic Minority Over‐Sampling Technique (SMOTE) was implemented to mitigate class imbalance. It generates synthetic samples in feature space by over‐sampling the minority class, leading to improving model performance in accurately classifying rare instances. Fine‐tuning the hyperparameters and implementing these improvements resulted in remarkable accuracy rates on a large BM dataset, with reduced training time and trainable parameters. CapsNet achieved 96.99%, VGG‐CapsNet achieved 98.95%, RES‐CapsNet achieved 99.24%, and the GN‐CapsNet model demonstrated superior accuracy at 99.45%. GN‐Caps Net was the best because it requires a small number of epochs and has an effective deep inception architecture that efficiently extracts features at different scales to form a robust representation of the input. Our proposed models were compared with existing state‐of‐the‐art models using the BM dataset; the results showed that our models outperformed the existing approaches and demonstrated excellent performance. Further, this automated system can analyze large amounts of data and complex cells in images of the BM dataset. Thus, it gives healthcare professionals a detailed understanding of different diseases, which may take time to achieve manually.

show abstract

A wavelet-based capsule neural network for ECG biometric identification

Cited by 20 publications

References 37 publications

Hybrid Deep Learning and Discrete Wavelet Transform-Based ECG Biometric Recognition for Arrhythmic Patients and Healthy Controls

Hybrid Deep Learning and Discrete Wavelet Transform-Based ECG Biometric Recognition for Arrhythmic Patients and Healthy Controls

ECG Biometrics Based on Attention Enhanced Domain Adaptive Feature Fusion Network

Examining the classification performance of pre‐trained capsule networks on imbalanced bone marrow cell dataset

Contact Info

Product

Resources

About