Dr. Mostafa Elsayed Ahmed Ibrahim :: Publications:

User authentication has become necessary in different life domains. Traditional authenti-13 cation methods like personal information numbers (PINs), password ID cards, and tokens are vul-14 nerable to attacks. For secure authentication, methods like biometrics have been developed in the 15 past. Biometric information is hard to lose, forget, duplicate, or share because it is a part of the hu-16 man body. Many authentication methods focused on electrocardiogram (ECG) signals have 17 achieved great success. In this paper, we have developed cardiac biometrics for human identifica-18 tion through a deep learning (DL) approach. Cardiac biometric systems rely on cardiac signals that 19 are captured through the electrocardiogram (ECG), photoplethysmogram (PPG), and phonocardio-20 gram (PCG). This study utilizes the ECG as a biometric modality because ECG signals are a superior 21 choice for accurate, secure, and reliable biometric-based human identification systems, setting them 22 apart from PPG and PCG approaches. To get better performance in terms of accuracy and compu-23 tational time, we have developed an ensemble approach based on VGG16 pre-trained transfer learn-24 ing (TL) and Long Short-Term Memory (LSTM) architectures to optimize features. To develop this 25 authentication system, we have fine-tuned this ensemble network. In the first phase, we prepro-26 cessed the ECG biosignal to remove noise. In the second phase, we converted the 1-D ECG signals 27 into a 2-D spectrogram image through a transformation phase. Next, the feature extraction step is 28 performed on spectrogram images through the proposed ensemble DL technique, and finally, those 29 features are identified by the boosting machine learning classifier to recognize humans. Several ex-30 periments were performed on the selected dataset, and on average, the proposed system achieved 31 98.7% accuracy, 98.01% precision, 97.1% recall, and 0.98 AUC. In this paper, we have compared the 32 developed approach with state-of-the-art biometric authentication systems. The experimental re-33 sults demonstrate that our proposed system outperformed the human recognition competition.