Academic & Scientific Articles

Permanent URI for this communityhttp://dl.cerist.dz/handle/CERIST/3

Browse

Has files: YesSubject: search.filters.subject.Deep learning

Search Results

Now showing 1 - 8 of 8
  • Thumbnail Image
    Item
    Privacy-preserving remote deep-learning-based inference under constrained client-side environment
    (Springer, 2023) Boulemtafes, Amine; Derhab, Abdelouahid; Ait Ali Braham, Nassim; Challal , Yacine
    Remote deep learning paradigm raises important privacy concerns related to clients sensitive data and deep learning models. However, dealing with such concerns may come at the expense of more client-side overhead, which does not fit applications relying on constrained environments. In this paper, we propose a privacy-preserving solution for deep-learning-based inference, which ensures effectiveness and privacy, while meeting efficiency requirements of constrained client-side environments. The solution adopts the non-colluding two-server architecture, which prevents accuracy loss as it avoids using approximation of activation functions, and copes with constrained client-side due to low overhead cost. The solution also ensures privacy by leveraging two reversible perturbation techniques in combination with paillier homomorphic encryption scheme. Client-side overhead evaluation compared to the conventional homomorphic encryption approach, achieves up to more than two thousands times improvement in terms of execution time, and up to more than thirty times improvement in terms of the transmitted data size.
  • Thumbnail Image
    Item
    Privacy-preserving deep learning for pervasive health monitoring: a study of environment requirements and existing solutions adequacy
    (Elsevier, 2022-03) Boulemtafes, Amine; Derhab, Abdelouahid; Challal , Yacine
    In recent years, deep learning in healthcare applications has attracted considerable attention from research community. They are deployed on powerful cloud infrastructures to process big health data. However, privacy issue arises when sensitive data are offloaded to the remote cloud. In this paper, we focus on pervasive health monitoring applications that allow anywhere and anytime monitoring of patients, such as heart diseases diagnosis, sleep apnea detection, and more recently, early detection of Covid-19. As pervasive health monitoring applications generally operate on constrained client-side environment, it is important to take into consideration these constraints when designing privacy-preserving solutions. This paper aims therefore to review the adequacy of existing privacy-preserving solutions for deep learning in pervasive health monitoring environment. To this end, we identify the privacy-preserving learning scenarios and their corresponding tasks and requirements. Furthermore, we define the evaluation criteria of the reviewed solutions, we discuss them, and highlight open issues for future research.
  • Thumbnail Image
    Item
    Deep learning in pervasive health monitoring, design goals, applications, and architectures: An overview and a brief synthesis
    (Elsevier, 2021-11) Boulemtafes, Amine; Khemissa, Hamza; Derki, Mohamed Saddek; Amira, Abdelouahab; Djedjig, Nabil
    The continuous growth of an aging population in some countries, and patients with chronic conditions needs the development of efficient solutions for healthcare. Pervasive Health Monitoring (PHM) is an important pervasive computing application that has the potential to provide patients with a high-quality medical service and enable quick-response alerting of critical conditions. To that end, PHM enables continuous and ubiquitous monitoring of patients' health and wellbeing using Internet of Things (IoT) technologies, such as wearables and ambient sensors. In recent years, deep learning (DL) has attracted a growing interest from the research community to improve PHM applications. In this paper, we discuss the state-of-the-art of DL-based PHM, through identifying, (1) the main PHM applications where DL is successful, (2) design goals and objectives of using DL in PHM, and (3) design notes including DL architectures and data preprocessing. Finally, main advantages, limitations and challenges of the adoption of DL in PHM are discussed.
  • Thumbnail Image
    Item
    Investigating Security and Privacy Concerns in Deep-Learning-based Pervasive Health Monitoring Architectures
    (IEEE, 2023-11) Boulemtafes, Amine; Amira, Abdelouahab; Derki, Mohamed Saddek; Hadjar, Samir
    Pervasive Health Monitoring (PHM) uses sensors and wearable devices and data analytics for real-time health monitoring. It enables early detection and personalized care interventions. This technology has the potential to revolutionize healthcare by improving proactive and preventive care.Besides, Deep learning (DL) based PHM is even more promising as it improves the discovery of complex patterns and correlations. This leads to precise health monitoring and personalized care, enhances diagnostics, and ultimately improves patient outcomes in the field of healthcare.However, privacy and security considerations must be addressed for successful implementation. This paper investigates the security and privacy concerns in Pervasive Health Monitoring architectures. It discusses through an illustrative DL-based PHM architecture the potential threats and attacks during the inference and training phases, and identifies key security and privacy issues. It also gives insights on countermeasures and technological solutions that can address security and privacy concerns in PHM architectures.
  • Thumbnail Image
    Item
    A new hybrid model of convolutional neural networks and hidden Markov chains for image classification
    (Springer, 2023-05) Goumiri, Soumia; Benboudjema , Dalila; Pieczynski, Wojciech
    Convolutional neural networks (CNNs) have lately proven to be extremely effective in image recognition. Besides CNN, hidden Markov chains (HMCs) are probabilistic models widely used in image processing. This paper presents a new hybrid model composed of both CNNs and HMCs. The CNN model is used for feature extraction and dimensionality reduction and the HMC model for classification. In the new model, named CNN-HMC, convolutional and pooling layers of the CNN model are applied to extract features maps. Also a Peano scan is applied to obtain several HMCs. Expectation–Maximization (EM) algorithm is used to estimate HMC’s parameters and to make the Bayesian Maximum Posterior Mode (MPM) classification method used unsupervised. The objective is to enhance the performances of the CNN models for the image classification task. To evaluate the performance of our proposal, it is compared to six models in two series of experiments. In the first series, we consider two CNN-HMC and compare them to two CNNs, 4Conv and Mini AlexNet, respectively. The results show that CNN-HMC model outperforms the classical CNN model, and significantly improves the accuracy of the Mini AlexNet. In the second series, it is compared to four models CNN-SVMs, CNN-LSTMs, CNN-RFs, and CNN-gcForests, which only differ from CNN-HMC by the second classification step. Based on five datasets and four metrics recall, precision, F1-score, and accuracy, results of these comparisons show again the interest of the proposed CNN-HMC. In particular, with a CNN model of 71% of accuracy, the CNN-HMC gives an accuracy ranging between 81.63% and 92.5%.
  • Thumbnail Image
    Item
    DIAG a diagnostic web application based on lung CT Scan images and deep learning
    (IOS Press Ebooks, 2021-05-29) Hadj Bouzid, Amel Imene; Yahiaoui, Saïd; Lounis, Anis; Berrani, Sid-Ahmed; Belbachir, Hacène; Naili, Qaid; Abdi, Mohamed El Hafedh; Bensalah, Kawthar; Belazzougui, Djamal
    Coronavirus disease is a pandemic that has infected millions of people around the world. Lung CT-scans are effective diagnostic tools, but radiologists can quickly become overwhelmed by the flow of infected patients. Therefore, automated image interpretation needs to be achieved. Deep learning (DL) can support critical medical tasks including diagnostics, and DL algorithms have successfully been applied to the classification and detection of many diseases. This work aims to use deep learning methods that can classify patients between Covid-19 positive and healthy patient. We collected 4 available datasets, and tested our convolutional neural networks (CNNs) on different distributions to investigate the generalizability of our models. In order to clearly explain the predictions, Grad-CAM and Fast-CAM visualization methods were used. Our approach reaches more than 92% accuracy on 2 different distributions. In addition, we propose a computer aided diagnosis web application for Covid-19 diagnosis. The results suggest that our proposed deep learning tool can be integrated to the Covid-19 detection process and be useful for a rapid patient management.
  • Thumbnail Image
    Item
    Ontology learning: Grand tour and challenges
    (Elsevier, 2021-02-21) Chérifa Khadir, Ahlem; Aliane, Hassina; Guessoum, Ahmed
    Ontologies are at the core of the semantic web. As knowledge bases, they are very useful resources for many artificial intelligence applications. Ontology learning, as a research area, proposes techniques to automate several tasks of the ontology construction process to simplify the tedious work of manually building ontologies. In this paper we present the state of the art of this field. Different classes of approaches are covered (linguistic, statistical, and machine learning), including some recent ones (deep-learning-based approaches). In addition, some relevant solutions (frameworks), which offer strategies and built-in methods for ontology learning, are presented. A descriptive summary is made to point out the capabilities of the different contributions based on criteria that have to do with the produced ontology components and the degree of automation. We also highlight the challenge of evaluating ontologies to make them reliable, since it is not a trivial task in this field; it actually represents a research area on its own. Finally, we identify some unresolved issues and open questions.
  • Thumbnail Image
    Item
    Side Channel Attack using Machine Learning
    (IEEE, 2022-12-15) Amrouche, Amina; Boubchir, Larbi; Yahiaoui, Saïd
    The overwhelming majority of significant security threats are hardware-based, where the attackers attempt to steal information straight from the hardware that our secure and encrypted software operates on. Unquestionably, side-channel attacks are one of the most severe risks to hardware security. Rather than depending on bugs in the program itself, a side-channel attack exploits information leaked from the program implementation in order to exfiltrate sensitive secret information such as cryptographic keys. A side channel assault could manifest in different ways including electromagnetic radiation, power consumption, timing data, or even acoustic emanation. Ever since the side-channel attacks were introduced in the 1990s, a number of significant attacks on cryptographic implementations utilizing side-channel analysis have emerged, such as template attacks, and attacks based on power analysis and electromagnetic analysis. However, Artificial Intelligence has become more prevalent. Attackers are now more interested in machine learning and deep learning technologies that enable them to interpret the extracted raw data. The aim of this paper is to highlight the main methods of machine learning and deep learning that are used in the most recent studies that deal with different types of side-channel attacks.