Publications Scientifiques
Permanent URI for this communityhttps://dspace.univ-boumerdes.dz/handle/123456789/10
Browse
13 results
Search Results
Item Classical and Quantum SVM for Electromyography-Based Myopathy Detection: A Comparative Exploration(Sciendo, 2025) Hammachi, Radhouane; Messaoudi, Noureddine; Belkacem, Samia; Pasetto, Edoardo; Delilbasic, AmerIntroduction: Electromyography (EMG) analysis is one of the most fundamental approaches for diagnosing neuromuscular diseases. Current advancements in technology have the potential to improve diagnosis accuracy using artificial intelligence (AI). Quantum machine learning (QML), while still in its early stages, offers promising potential for various medical applications, but its effectiveness in real-world diagnostic tasks needs further exploration. Thus, the aim of this study is to employ both quantum and classical support vector machines (SVMs) to classify EMG signals into two classes, healthy and myopathy, and compare their performance. Methods: Various approaches were tested; classical SVM and quantum-kernel-based SVM, both with manually extracted features, and convolutional neural network (CNN)-based deep features extraction techniques. This allows for an evaluation of the strengths and limitations of this new technology, acknowledging the potential of both classical and quantum methods. Results: The obtained results showed that the proposed quantum methods yielded promising outcomes and comparable to classical methods. Particularly, the competitive results of the quantum SVM (QSVM) with the CNN-based deep feature extraction approach, which delivered a high training and testing accuracies of up to 96.7% and 85.1%, respectively. Conclusion: These findings encourages the necessity for more advanced QML research, particularly in medical applications as quantum technology progressesItem Using Modified Gorti-enhanced Homomorphic Cryptosystem to Improve Security of ECG Signal(National Institute of Telecommunications, 2025) Besmi, Fatma Zohra; Belkacem, Samia; Messaoudi, NoureddineWhile offering vast data storage capabilities, cloud computing poses numerous security- and privacy-related challenges. This requires robust security measures, particularly for sensitive data, such as electrocardiograms (ECG). Homomorphic encryption (HE) emerges as a promising solution by enabling secure computations to be performed directly on encrypted data. This study introduces a novel approach to enhance the security of ECG data. We modified the Gorti-enhanced homomorphic cryptosystem (MEHC) method by optimizing its key generation procedure and then applied the linear congruential generator (LCG) algorithm to create a list of huge prime integers. Furthermore, we increased the modulus value and enlarged the message space. These enhancements boosted overall security by substantially improving immunity to factorization attacks. We used quantization and fixed-point representation to enhance the encryption process. As an additional security layer, an evaluation process has been added to the proposed algorithm which performs various mathematical operations homomorphically on the encrypted data, rather than on the original data. This modified algorithm enables efficient and secure encryption of ECG data while preserving the ability to reliably identify arrhythmias, such as bradycardia and tachycardia. Using the MIT-BIH arrhythmia database, the proposed MEHC system demonstrated high accuracy (98.48%), sensitivity (99.10%) and positive predictive value (99.33%), while effectively safeguarding the ECG data. These results validate the efficacy of the MEHC system and confirm its suitability for secure and reliable ECG signal processing in healthcare applicationsItem Classical and Quantum SVM for Electromyography-Based Myopathy Detection: A Comparative Exploration(Polish Society of Medical Physics, 2025) Hammachi, Radhouane; Messaoudi, Noureddine; Belkacem, Samia; Pasetto, Edoardo; Delilbasic, AmerIntroduction: Electromyography (EMG) analysis is one of the most fundamental approaches for diagnosing neuromuscular diseases. Current advancements in technology have the potential to improve diagnosis accuracy using artificial intelligence (AI). Quantum machine learning (QML), while still in its early stages, offers promising potential for various medical applications, but its effectiveness in real-world diagnostic tasks needs further exploration. Thus, the aim of this study is to employ both quantum and classical support vector machines (SVMs) to classify EMG signals into two classes, healthy and myopathy, and compare their performance. Methods: Various approaches were tested; classical SVM and quantum-kernel-based SVM, both with manually extracted features, and convolutional neural network (CNN)-based deep features extraction techniques. This allows for an evaluation of the strengths and limitations of this new technology, acknowledging the potential of both classical and quantum methods. Results: The obtained results showed that the proposed quantum methods yielded promising outcomes and comparable to classical methods. Particularly, the competitive results of the quantum SVM (QSVM) with the CNN-based deep feature extraction approach, which delivered a high training and testing accuracies of up to 96.7% and 85.1%, respectively. Conclusion: These findings encourages the necessity for more advanced QML research, particularly in medical applications as quantum technology progresses.Item Deep Learning Classification of Simulated Surface EMG Signals across Maximum Voluntary Contraction Levels(Institute of Biophysics and Biomedical Engineering at the Bulgarian Academy of Sciences, 2025) Hammachi, Radhouane; Belkacem, Samia; Messaoudi, Noureddine; Bekka, Raïs El’hadiElectromyography (EMG) is a fundamental tool in diagnosing neuromuscular disorders (NMD). Due to the complex nature of EMG signals, different approaches, based on artificial intelligence and machine learning, were developed for EMG signal analysis and NMD diagnosis. Considering the critical role of maximum voluntary contraction (MVC) as a fundamental metric in assessing muscle fatigue, in this work, classification of simulated surface EMG (sEMG) into MVC levels is performed. Unlike previous studies, which focus primarily on binary classification of fatigue and non-fatigue states, our approach employs a deep convolutional neural network for the classification of sEMG signals into ten MVC levels, where the model outputs categorical predictions, with each class representing a specific MVC level. sEMG signals were generated using a computer muscle model that we developed using MATLAB, which allows for greater control over variability, ensuring robustness and generalizability of the model. The obtained results demonstrate that the model achieved high performance in differentiating between the ten classes (MVC levels), with an accuracy, F1-score, recall, and precision of 88.88%, 88.75%, 88.80% and 88.86%, respectively. These findings reveal that the model can accurately differentiate across MVC levels, indicating a potential method for accurate assessment of muscle fatigue intensity.Item Efficient invisible color image watermarking based on chaos(Institute of Advanced Engineering and Science (IAES), 2024) Belkacem, Samia; Messaoudi, NoureddineSeveral difficulties are faced in developing a robust and transparent color image watermarking system, which requires the blending of the human visual system (HVS) during its design. Therefore, employing masks that take into account the features of HVSs has become a very effective tool for boosting robustness requirements without significant alterations in image imperceptibility. The present article offers watermarking strategy for colored images employing a reverse self-reference image in conjunction with the HVS constraint. A color image first undergoes conversion through the Red, Green, and Blue (RGB) format to the National Television Systems Committee (NTSC) space. The reference image is derived from the luminance channel through the discrete wavelet transform (DWT) domain. However, the chaotic map serves to generate the watermark, and a 2D torus automorphism is subsequently used to scramble the watermark. Therefore, the watermark is scrambled and placed in the reference image. Moreover, the detecting phase involves the host image, where the reference image is extracted from both the host and the image with a watermark, and the correlation is subsequently used to assess the similarity between the retrieved and the introduced watermark. The proposed watermarking scheme can retain the watermarked image's perceptibility justified by the PSNR. In addition, it achieves high robustness to withstand a wide array of attacks.Item Randomness evaluation of coupled chaotic maps via NIST tests: A comparative study(IEEE, 2020) Bourekouche, Hadjer; Belkacem, Samia; Messaoudi, NoureddineA vital requirement for any random number generator based on chaos is to ensure that the generated sequence always benefits of a significant level of randomness. It is critical to examine such sequences by means of Lyapunov exponents, bifur-cation diagrams, or other tests in order to accurately select the parameters of the dynamic system. However, the sequence’s randomness quality varies depending on the generator's design and must be examined in different ways. Therefore, we argue to use the National Institute of Standards and Technology (NIST) suite tests to evaluate and compare the randomness properties of two coupled systems found in existing literature: the logistic-sine system (LSS) and the logistic-tent system (LTS). The results reveal that the LSS has much superior statistical features in terms of randomness than the LTS in the range [3.1–4]. This conclusion will substantially affect the selection of the perfect chaotic map to create sequences of keys that match the requirements of cryptography applications.Item Efficient image encryption scheme using a nonlinear shift register and chaos(TARU PUBLICATIONS, 2024) Bourekouche, Hadjer; Belkacem, Samia; Messaoudi, NoureddinePowerful cryptographic systems require a qualified random number generator. This research purposes to provide a comprehensive comparative analysis done on several of the well-known pseudo-random number generators (PRNGs) regarding their efficiency and resilience against crypto-analytical threats. These generators consist of the basic 8-bit Non- Linear Feedback Shift Register (NLFSR), the logistic map (LM), and our proposed hybrid random number generator named NLFSR-LM, which combines through XOR operation the sequences of the NLFSR with the LM to achieve a high quality of randomness. The performance of the created generator is examined and subsequently compared according to statistical tests of randomness alongside cryptographic features in terms of key space, key sensitivity and resistance to numerous attacks. The proposed generator produced good results and exhibited several interesting properties, such as a high degree of security, a sufficiently large key space, and it provided better randomness than other frequently used PRNGs.Item Simulated surface electromyographic (semg) signal generation and detection model(Sciendo, 2023) Messaoudi, Noureddine; Belkacem, Samia; Bekka, Rais El’hadiFor didactic purposes, the aim of this work was to improve a simulation model of surface electromyographic (sEMG) signal by taking into consideration the shortcomings of previously developed models. This model started with the simulation of the single fibre action potential (SFAP), then the model of the single motor unit action potential (MUAP), afterwards the imitation of the train of MUAP and finally the modellig of the resultant sEMG signal which is the sum of the MUAPs trains. SFAP simulation was based on: i) the description of the volume conductor model which is composed of four layers (bone, muscle, fat and skin), ii) the description of the electrodes shapes and sizes as well as spatial filters, iii) and the transmebrane current. The proposed model shows its effectiveness in the possibility of carrying out practical work by simulation on the modelling of SFAP, MUAP, MUAPT and the sEMG signal. The most important result of this model is that signal processing tools can be applied to analyze and interpret real-world phenomena such as the effects of physiological , non physiological and sensing system parameters on the shape of the simulated sEMG signal.Item ECG beats classification with interpretability(IEEE, 2022) Hammachi, Radhouane; Messaoudi, Noureddine; Belkacem, SamiaRecently, a lot of emphasis has been placed on Artificial Intelligence (AI) and Machine Learning (ML) algorithms in medicine and the healthcare industry. Cardiovascular disease (CVD), is one of the most common causes of death globally, and Electrocardiogram (ECG) is the most widely used diagnostic tool to investigate this disease. However, the analysis of ECG signals is a very difficult process. Therefore, in this work, automated classification of ECG data into five different arrhythmia classes is proposed, based on MIT-BIH dataset. Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) Deep Learning (DL) models were used. The black-box nature of these complex models imposes the need to explain their outcomes. Hence, both Permutation Feature Importance (PFI) with Gradient-Weighted Class Activation Maps (Grad-CAM) interpretability techniques were investigated. Using the K-Fold cross-validation method, the models achieved an accuracy of 97.1% and 98.5% for CNN and LSTM, respectivelyItem Ability of spatial filters to distinguish between two MUAPs generated from MUs with different locations, sizes and fibers pennation(IOP Publishing, 2023) Messaoudi, Noureddine; Belkacem, Samia; Bekka, Raïs El’hadiIn this study, we investigated the effects of the motor unit (MU) location and size and the fibres pennation on the ability of anisotropic and almost isotropic spatial filters used to detect surface electromyographic (EMG) signals to make a distinction between motor unit action potentials (MUAPs) generated from two MUs. The study was based on simulated MUAPs. The fibres orientation was performed by varying the fibres pennation angle (FPA). The root mean square error (RMSE) between MUAPs generated from two MUs was used as a criterion to evaluate the ability of the investigated filters to distinguish between two generated MUAPs. The location of a MU was fixed and the second MU moved away from the first MU in the transversal direction for the first case and in the depth direction in the second case to take five different locations in every case. We showed that the capability of the studied filters to more separate two MUAPs strongly depended on MU location, MU size and FPA. This capability of separation was best with large distances between the two MUs and with large sizes of them. Furthermore, the main survey of this work was that the BiTDD filter has the best ability of separation of two MUAPs than the other filters in a given FPA interval. The number of pennation angles in this interval is related to the location and size of the moved MU