Deep neural networks play a significant role in medical image analysis, particularly in improving the efficiency and accuracy of disease diagnosis and treatment planning. The ability to preserve the privacy of medical data opens the door to harnessing more information to train powerful and intelligent AI models. However, the added complexity introduced by incorporating privacy measures often leads to increasingly opaque deep learning models. As the need to interpret these privacy-preserving deep learning models grows, explainable AI systems have become pivotal for cultivating trust among clinical experts and stakeholders. To address this challenge, researchers have begun to focus on developing privacy-preserving techniques with improved explainability features. This article presents a comprehensive survey of different privacy models and security techniques, focusing on biomedical imaging applications. The survey covers peer-reviewed studies published during 2019-2024, ensuring comprehensive and contemporary coverage of privacy-preserving and explainability techniques. It covers a detailed review of various privacy-preserving methods and studies related to model explanations. It also highlights unresolved challenges and suggests potential research directions. This survey aims to offer valuable directions to the research community by explaining privacy-preserving techniques for medical image analysis.

The increasing number of international students (IS) enrolled in Australian higher education institutions, combined with the widespread adoption of online and hybrid learning, has significant implications for understanding the factors that influence engagement among this diverse student group. Early identification of students with low engagement facilitates academic success, prevents poor outcomes, optimises resource allocation, improves teaching strategies, increases motivation, and supports long term success. . This study's main aim is to examine the use of AI to predict student engagement. Development of a theoretically informed survey that aimed to elicit post graduate students' engagement was developed and validated by expert judgement. In total, 200 copies of the survey were distributed, 121 responses were received, and 96 were considered for this study representing a response rate of 48%. This study promotes a multidimensional approach, utilising AI and ML methodologies, to determine the influence of social and cultural contexts on student engagement This approach enables educators and institutions to create effective strategies for enhancing the learning experience of postgraduate students. Multiple AI and ML techniques have been utilised including synthetic data generation methods such GaussianCopula, TVAE, GAN, CopulaGAN, and CTGAN. These techniques are specifically employed to predict various dimensions of engagement, including personal, academic, intellectual, social, and professional engagement. . The performance of AI/ML algorithms, including SVM, KNN, DT, GBM, RF, NB, LR, and ET, was assessed using several metrics including F1 Score, Sensitivity, Specificity, Confusion Matrix, and Accuracy. The models used in this study achieved up to 85% accuracy, offering a solid foundation for guidelines and support to enhance decision making processes in higher education. These findings provide valuable insights for both academics and policy makers, laying the groundwork for evidence-based strategies to improve student engagement.

This paper addresses the challenge of fault analysis in the Vehicular Energy Network (VEN) caused by small fault samples due to transient faults and complex disturbances. The proposed generation and diagnosis networks (GDNs) are developed without necessitating prior knowledge or manual intervention. The approach starts with an encoding and diagnosis network that converts multi-dimensional signals into images through supervised learning. A sample enhancement network, improved with module transfer and a relaxation objective function, is then proposed to increase the reliability of convergence and diversity of features for small fault samples. Additionally, a joint iterative training strategy between these two networks improves diagnostic accuracy and generalization through feature feedback. Performance validation on a semi-physical simulation platform demonstrates that the proposed GDNs achieve a 20% improvement in diagnostic accuracy with small datasets (200 samples) and maintain superior performance as sample volume grows. Thus, the proposed approach offers a potent solution for fault diagnosis in VENs with scarce samples, enhancing the analysis of complex systems. Note to Practitioners-This paper delves into fault diagnosis in the vehicular energy network (VEN) using small samples, employing a data-driven and deep learning model. The proposed method is versatile, suitable for analyzing complex systems with multiple and heterogeneous signals. An end-to-end model, named generation and diagnosis networks (GDNs), is introduced for generating small samples and conducting fault diagnosis without requiring prior knowledge or manual input. This method encodes multiple signals into signal images, which are then processed by a specially designed sample enhancement model, improved through the relaxation objective function and module transfer method. The enhanced samples are utilized for accurate analysis within the encoding and diagnosis networks' diagnostic unit. The paper also provides a comparison of diagnosis results for reference. This approach enables researchers and engineers to efficiently augment and analyze small samples in practical applications, offering a practical framework for junior and inexperienced analysts. Preliminary experiments conducted using the RT-Lab semi-physical simulation platform suggest the method's feasibility and effectiveness. Future research will explore the optimization of the model's topology and parameters for lightweight design.

Deep learning for plant disease recognition faces challenges in real-world scenarios, especially on resource-constrained devices and with complex backgrounds. While deep neural networks excel with such images, lightweight approaches struggle due to the complexities in feature extraction, affecting classification accuracy. To overcome this, we introduce a novel pipeline that masks and removes complex backgrounds, enabling accurate disease recognition from affected leaves. This approach also significantly increases model accuracy and reduces training time compared to baseline models due to object-cropped input. Using various lightweight neural networks as backbones for the classification task, we validated the efficiency of our approach, which outperformed the baseline in terms of performance. Specifically, our proposed background masking model achieved an IoU of 93.93% for background masking. Furthermore, for disease recognition, our approach using the SqueezeNet-1.1 backbone demonstrated the highest accuracy of 90.84%, surpassing the baseline’s highest average accuracy of 89.22%. These results demonstrate the efficacy of our proposed method in enhancing disease recognition in agricultural imagery, thereby contributing to the advancement of deep learning techniques in this domain. Moreover, the comparison of training times demonstrates that our approach achieved a considerable reduction compared to the benchmark.

Technological evolution in the Industrial Internet of Things (IIoT) domain has fostered smart grid systems' operation, performance, connectivity, and delivery with higher efficiency. However, it has also exposed the platform to a broader surface for attackers. Current information technology (IT)-centric solutions for detecting, preventing, and mitigating attacks have limitations, especially in comprehensively monitoring industrial control operational technology (OT) and communication systems. The rise of sophisticated cyberattacks, such as targeted ransomware, demand more robust security measures, leading to the emergence of zero trust (ZT) deployment as a response to these threats. This article proposes a new framework for implementing ZT comprising both IT and OT in smart grid infrastructures, with multiple security mechanisms and robust system coverage. We present an EigenGame algorithm for integrating diverse data sources into a rich-context format and an enhanced approach to quantum reinforcement learning for reliable malicious behavior detection in IIoT-enabled smart grids. The framework was evaluated using five sets of data from the X-IIoTID dataset, demonstrating its good performance in verifying any behavior inside the system and identifying any malicious behavior related ransomware attacks.

In the era of zero trust security models and next-generation networks (NGN), the primary challenge is that network nodes may be untrusted, even if they have been verified, necessitating continuous validation and scrutiny. Effective intrusion detection systems (IDS) are crucial for continuously monitoring network traffic and identifying potential threats. However, traditional IDS approaches often struggle to keep pace with evolving threats, requiring extensive supervised training on labeled datasets. This limitation leads to high false positive rates, low detection accuracy, and a failure to provide real-time detection, thereby undermining the security of NGNs. This paper proposed the first self-supervised learning-based IDS, designed on temporal contrastive graph neural network (GNN), namely TCG-IDS . It innovatively integrates three contrastive learning strategies: temporal contrasting to capture temporal dependencies, asymmetric contrasting to account for the diverse interactions within network data, and masked contrasting to enhance the learning of node representations by masking parts of the data during training. Performance evaluation was conducted on two publicly available network traffic datasets, NF-CSE-CIC-IDS2018-V2 and NF-UNSW-NB15-V2. TCG-IDS achieved a balanced accuracy of 99.48% and 91.48% on two datasets respectively, significantly outperforming state-of-the-art graph learning models. In multi-class detection, TCG-IDS attained a mean false positive rate of 4.15% and 3.34% on the two datasets respectively. Besides, it exhibits high efficiency with its running time of 0.37s and 0.51s on the two datasets to predict per batch of 100 samples. Results highlight the effectiveness and efficiency of TCG-IDS in accurately detecting various types of network intrusions. This work significantly advances the field of network intrusion detection via self-supervised temporal graph learning, offering a promising solution for future network security systems.

Smart Parking Services (SPSs) enable cruising drivers to find the nearest parking lot with available spots, reducing the traveling time, gas, and traffic congestion. However, drivers risk the exposure of sensitive location data during parking query to an untrusted Smart Parking Service Provider (SPSP). Our motivation arises from a repetitive query to an updated database, i.e., how a driver can be repetitively paired with a previously-matched-but-forgotten lot. Meanwhile, we aim to achieve repetitive query in an oblivious and unlinkable manner. In this work, we present Mnemosyne2 : decentralized and privacy-preserving smart parking with secure repetition and full verifiability. Specifically, we design repetitive, oblivious, and unlinkable Secure k Nearest Neighbor (SkNN) with basic verifiability (correctness and completeness) for encrypted-andupdated databases. We build a local Ethereum blockchain to perform driver-lot matching via smart contracts. To adapt to the lot count update, we resort to the immutable blockchain for advanced verifiability (truthfulness). Last, we utilize decentralized blacklistable anonymous credentials to guarantee identity privacy. Finally, we formally define and prove privacy and security. We conduct extensive experiments over a real-world dataset and compare Mnemosyne2 with existing work. The results show that a query only needs 8 seconds (175 ms) on average for service waiting (verification) among 500 drivers.

The Industrial Internet of Things (IIoT) is a collection of interconnected smart sensors and actuators with industrial software tools and applications. IIoT aims to enhance manufacturing and industrial processes by capturing and analyzing real-time industrial data. However, the heterogeneous and homogeneous nature of IIoT networks makes them vulnerable to several security threats. As data is transmitted over an insecure communication medium, intruders may intercept communication among different entities and perform malicious activities. Consequently, ensuring the security and privacy of data transmitted in IIoT networks is essential. Motivated by the aforementioned challenges, this article presents a deep-learning-integrated blockchain framework for securing IIoT networks. Specifically, first, we design a private blockchain-based secure communication among the IIoT entities using session-based mutual authentication and key agreement mechanism. In this approach, the Proof-of-Authority (PoA) consensus mechanism is used for verification of the transactions and block creation based on the voting of miners over the cloud server. Second, we design a novel deep-learning-based intrusion detection system that combines contractive sparse autoencoder (CSAE), attention-based bidirectional long short-term memory (ABiLSTM) networks, and softmax classifier for cyberattack detection. The practical implementation of blockchain and deep-learning techniques proves the effectiveness of the proposed framework.

Due to the complexity and diversity of Industrial Internet of Things (IIoT) systems, which include heterogeneous devices, legacy and new connectivity protocols and systems, and distributed networks, sophisticated attacks like ransomware will likely target these systems in the near future. Researchers have focused on studying and addressing ransomware attacks against various platforms in recent years. However, to the best of our knowledge, no existing study investigates the new trends of ransomware tactics and techniques and provides a comprehensive analysis of ransomware attacks and their detection techniques for IIoT systems. Therefore, this paper investigates this attack and its associated detection techniques in IIoT systems in various aspects, including recent ransomware tactics, types, infected operating systems, and platforms. Specifically, we initially discuss the evolution of the IIoT system and its common architecture. Then, we provide an in-depth examination of the development of ransomware attacks and their constituent blocks, outline recent tactics and types of ransomware, and provide an extensive overview of the latest research on detection models. We also summarize numerous significant issues that have yet to be addressed and require further research. We conclude that offensive and defensive research is urgently needed to protect IIoT against ransomware attacks.

In today’s cybersecurity landscape, software security companies encounter a significant challenge in detecting new and unknown malware. Despite the introduction of various machine learning and deep learning tools designed to identify malicious software based on static and dynamic features, achieving the desired level of accuracy remains elusive. This challenge is exacerbated by factors such as encryption, packing, limited distribution, and uneven allocation of malware samples across different families. Moreover, deep learning techniques demand substantial time, computational resources (specifically GPUs), and expertise from data scientists for practical malware analysis. In response to these challenges, we propose a novel GPU-free approach called Image-based Malware Classification using Broad Learning (IMCBL) to address these issues. Our method integrates visualization, feature decomposition, and broad learning architecture to enhance malware detection and classification. We convert raw malware binaries into images, reducing the necessity for extensive feature engineering. These images transform using truncated Singular Value Decomposition (SVD) to reduce the feature vector size, expediting the training process while mitigating model overfitting. The transformed feature vector is then input into our proposed Broad Learning (BL) system, which facilitates malware detection and classification. The BL architecture, structured as a flat network mapping original inputs to feature nodes and expanding the structure in enhancement nodes, ensures efficient and effective classification without the need for retraining. This dynamic and incremental learning capability sets IMCBL apart, making it superior to existing deep learning architectures. To validate our approach, we conducted extensive experiments using five benchmark malware datasets, including the Microsoft Windows malware challenge dataset, the Malimg Windows malware dataset, the IoT-Android mobile malware dataset, the Big Windows malware dataset, and an obfuscated Windows malware dataset. The results demonstrate IMCBL’s remarkable success in classifying most malware samples, even under obfuscation attacks, performing comparably or outperforming current methods using similar benchmarks. Specifically, IMCBL achieved 95.58% accuracy for the Microsoft Windows malware dataset, 97.64% accuracy for the Malimg Windows malware dataset, 96.51% accuracy for IoT Android malware datasets, and 96.19% accuracy for the extensive Windows malware dataset. Additionally, IMCBL demonstrated 93.04% accuracy for an obfuscated Windows malware dataset, which contains both packed and unpacked malware samples. Notably, IMCBL exhibits an exponential advantage in computation overhead, including training and prediction time, when compared to traditional machine learning and advanced state-of-the-art deep learning architectures such as VGG16, ResNet50, and InceptionV3.

The abnormal growth of malignant or nonmalignant tissues in the brain causes long-term damage to the brain. Magnetic resonance imaging (MRI) is one of the most common methods of detecting brain tumors. To determine whether a patient has a brain tumor, MRI filters are physically examined by experts after they are received. It is possible for MRI images examined by different specialists to produce inconsistent results since professionals formulate evaluations differently. Furthermore, merely identifying a tumor is not enough. To begin treatment as soon as possible, it is equally important to determine the type of tumor the patient has. In this paper, we consider the multiclass classification of brain tumors since significant work has been done on binary classification. In order to detect tumors faster, more unbiased, and reliably, we investigated the performance of several deep learning (DL) architectures including Visual Geometry Group 16 (VGG16), InceptionV3, VGG19, ResNet50, InceptionResNetV2, and Xception. Following this, we propose a transfer learning(TL) based multiclass classification model called IVX16 based on the three best-performing TL models. We use a dataset consisting of a total of 3264 images. Through extensive experiments, we achieve peak accuracy of 95.11%, 93.88%, 94.19%, 93.88%, 93.58%, 94.5%, and 96.94% for VGG16, InceptionV3, VGG19, ResNet50, InceptionResNetV2, Xception, and IVX16, respectively. Furthermore, we use Explainable AI to evaluate the performance and validity of each DL model and implement recently introduced Vison Transformer (ViT) models and compare their obtained output with the TL and ensemble model.

The Internet of Things (IoT) has recently received a lot of attention from the information and communication technology community. It has turned out to be a crucial development for harnessing the incredible power of wireless media in the real world. The nature of IoT-Fog networks requires the use of defense techniques who are light and mobile-aware. The edge resources in such a distributed environment are open to various safety hazards. DDoS UDP flooding attacks are the most frequent threats to edge resources in IoT-Fog networks. It is crucial for sabotaging fog gateways and can overcome traditional data filtering techniques. This paper introduces M-RL, a lightweight intrusion detection system with mobility awareness that can detect DDoS UDP flooding attacks while taking into account adversarial IoT devices that engage in IP spoofing. To this end, this paper analyzes the malicious behaviors that result in anonymity against Rate Limiting and Received Signal Strength (RSS)-based approaches, combines their advantages, and addresses their vulnerabilities. We test our method in different contexts to achieve that goal, and we find that it may decrease the accuracy of the RL, RSS, and RSS-RL methods to 70%, 48.9%, and 64.3%, respectively. The outcomes demonstrate the proposed approach's resistance to software-based source address forgery, impersonation, and signal modification. It offers more than 99% accuracy and supports node mobility. In this case, the best possible accuracy of the previous methods is 77%.

Range queries allow data users to outsource their data to a Cloud Server (CS) that responds to data users who submit a request with range conditions. However, security concerns hinder the wide-scale adoption. Existing works neglect item availability, fail to protect secure verification or sacrifice search accuracy for efficiency. In this paper, we propose Secure, Available, Verifiable, and Efficient (SAVE) range query processing, which has three distinctive features. (1) Secure availability checking against a malicious CS: we design a keyed index-based secure verification mechanism to check the availability of matched nodes, including validity and freshness. (2) Secure result verification: we design a targeted verification mechanism for result correctness and completeness while not compromising security. (3) Improved efficiency and accuracy: we design a layered encoding method to improve search efficiency and accuracy. We formally stated and proved the security of SAVE in the random oracle model. We conducted extensive experiments over the Yelp and FourSquare dataset to validate the efficiency, e.g., a query over 10 thousand data items only needs 19.4 ms to get queried results and 3.5 ms for local verification.

The successful deployment of an Intrusion Detection System (IDS) in the Internet of Things (IoT) is subject to two primary criteria: the detection method and the deployment strategy. IDS schemes should take into account that IoT devices often have limited resources. Thus, IDS should be limited in devices’ memory and power usage. In this paper, we design, implement, and evaluate an effective cross-layer lightweight IDS scheme for the IoT (RPL-IDS). The proposed IDS scheme cooperates with the RPL routing protocol using its selected parents as distributed agents. A lightweight artificial neural network (ANN) model is deployed in these agents to detect malicious traffic and collaborates with a centralized system. According to the topology built by the Routing Protocol for Low-Power and Lossy Networks (RPL), these agents are automatically selected, i.e., the routers (parents) of the topology are chosen to act as IDS agents. We implemented RPL-IDS using the Contiki operating system and then comprehensively evaluated it with the Cooja simulator. Experimental results indicate that RPL-IDS is lightweight and can be deployed on devices with limited resources. Most state-of-the-art IDS schemes do not consider the limitation of resources of IoT devices, making them impractical for deployment in many IoT applications. Furthermore, the proposed RPL-IDS demonstrated one of the highest detection rates in the literature while incurring an insignificant energy overload, allowing for scalability in large-scale networks.

Applying artificial intelligence (AI) to data from Industrial Internet of Things (IIoT) devices is a novel direction in geological studies. However, privacy and security concerns hinder the sharing of data, thus affecting the performance of current AI-based approaches. In this article, we propose a novel data management style to address the privacy and security issues in joint hydrocarbon explorations. Federated learning can facilitate the analysis of multiple datasets without the need to share them, protecting private information of different companies in a virtual joint venture. We use the inference of petroleum reservoirs in karst stratigraphy as a case study. A federated learning-based enterprise data management framework is proposed to virtually integrate the information from different organizations. Our key contributions are summarized as follows. 1) A method for karst identification and inference is proposed, which uses neural networks to recognize the size of petroleum reservoirs in different karst areas. 2) A federated learning algorithm is applied to virtually aggregate data samples from different companies. 3) The performance of the new privacy-preserving integration model is compared with those of the individual/local deep learning models. Our results show that the proposed approach can substantially improve the accuracy of petroleum reservoir explorations.

Extreme events (such as earthquakes, hurricanes, etc.) pose a dual challenge to the reliability and serviceability of IoT systems. With regard to this challenge, by publishing some tasks and then encouraging the public to assist in real-time data collection through their mobile terminals (namely, the mobile crowdsourcing-assisted IoT systems), is expected to play an important role in secondary disaster prevention and personnel rescue in extreme events. However, it has weaknesses in terms of security, flexibility, and efficiency. As an elegant solution, identity-based broadcast proxy re-encryption (PR-IBBE) enables flexible access authorization sharing and efficient broadcast distribution of encrypted tasks via the cloud. However, their security relies on fully-trusted or semi-trusted cloud assumptions, which are hard to be implemented in real-world scenarios. And the cloud is more vulnerable in an emergency event since there is a lack of effective management. Motivated by that, we propose the verifiable PR-IBBE (VPR-IBBE) scheme, which realizes a cross-domain identity-based broadcast task file secure authorization access, and empowers the verifiability and reputability of re-encrypted ciphertext under the untrusted cloud setting. This mechanism ensures that the relevance between the re-encrypted ciphertext and the original ciphertext can be publically verified, so the cloud can defend itself if there is a malicious accusation of forging the re-encrypted ciphertext. Through rigorous formal security proofs, we demonstrate that VPR-IBBE attains the indistinguishability of ciphertext against selective identity chosen ciphertext attack (IND-sID-CPA), and is also resistant to the collusion attack between the untrusted cloud and the cooperative performer. Theoretical comparison and experimental results demonstrate the practicability of our VPR-IBBE scheme, as well as the superiority over representative related works.