The Internet of Medical Things (IoMT) has significantly enhanced the healthcare sector by enabling advanced connectivity between smart medical devices, improving patient monitoring, and optimizing care quality. However, this increasing connectivity exposes these systems to cyberattacks that can compromise data confidentiality, disrupt device operations and endanger patient safety. Attacks on the IoMT, such as intrusions, ransomware, data tampering and denial-of-service, exploit vulnerabilities in medical devices, which often lack adequate security measures. Addressing these challenges requires robust attack detection solutions tailored to the constraints of IoMT environments. In this paper, we propose a novel approach based on Federated Learning (FL), enabling medical devices to collaboratively detect cyberattacks without directly sharing sensitive data. Unlike traditional FL systems that rely on computationally intensive neural networks, our approach leverages XGBoost—a lightweight yet powerful algorithm—to train detection models locally on devices. The XGBoost models are further optimized using a Bayesian method and integrated with an aggregation algorithm to construct an adaptive global model. We evaluate our approach using three IoMT datasets: CICIoMT2024, ECU-IoHT, and WUSTL-EHMS. Experimental results demonstrate that, in addition to minimizing the risk of medical data leakage, our method achieves superior detection accuracy compared to centralized approaches.

The early treatment and detection of plant diseases are important processes, as many diseases affecting crops are highly contagious. Recent advancements in deep learning have helped to provide innovative tools that have not only assisted early detection, but also significantly improved the performance and accuracy of Coffee Leaf Disease (CLD) classification and treatment. However, training a deep learning model from scratch can be both resource and time-consuming. To overcome this challenge, the transfer learning technique can take full advantage of pre-trained models for more general tasks on extensive datasets o ameliorate the performance of a new, related task using few-shot training. This paper proposes a deep learning model, coupled with transfer learning, for CLD detection that aims to provide high accuracy forecasting of diseases that could affect coffee leaves. Our method involves 195 different pre-trained deep learning models, including real-time models like MobileNet and dense ones like EfficientNet and ResNet for the detection of four different diseases. The findings suggest that the EfficientNetB0 model, with transfer learning, has the most relevent accuracy (99.99%), and thus offers an effective solution for coffee leaf diseases classification of. This result could be used to develop applications that help coffee growers to improve the productivity and quality of their crops through early and accurate detection of coffee plant leaf diseases. Such an Artificial Intelligence based application would provide growers with timely control measures, preventing the spread of disease, and minimizing crop damage.

Mobile computing as ubiquitous and pervasive technology supports portable and context-aware computation. To date, there exist a significant number of traditional computing...

In recent years, several research and development initiatives have focused on developing secure and trustworthy systems for the healthcare industry via pervasive and mobile healthcare (mHealth) solutions. State-of-the-art mHealth solutions primarily rely on centralized storage, such as cloud computing servers, which may escalate the maintenance costs, require ever-increasing storage infrastructure, and pose privacy and security risks to the health-critical data produced, consumed, and transmitted over ad hoc networks. To overcome these limitations, we conducted this study intending to synergize mobile computing (devices to process health-critical data) and blockchain technology (infrastructure to secure storage and retrieval of health-critical data), specifically addressing data security and privacy using a blockchain mHealth system. The research employs an incremental method by (i) developing a framework that acts as a blueprint to architect blockchain-enabled mHealth systems, (ii) implementing a suite of algorithms as a proof-of-concept to automate the framework, and (iii) experimental evaluations to validate the scalability, computation, and energy efficiency of the proposed solution. The proposed framework has been implemented as a frontend using a mobile application interface that exploits the backend via the InterPlanetary File System (IPFS) system and Ethereum blockchain for secure management of mHealth data. We use a case-study-based approach demonstrating how health units, medics, and patients can securely access and distribute health-critical data. For evaluation, we deployed a smart contract prototype on the Ethereum TESTNET network in a Windows environment to test the proposed framework. Results of the evaluation indicate (a) scalability with query response time (range: 10–41 ms), (b) computational performance (CPU utilization: 1.5% – 2.5%), and (c) energy efficiency (gas consumption: 40000 units for 1000 bytes). The proposed solution – framework, algorithms, and experimental evaluation – aims to advance state-of-the-art architecting and implementing cybersecurity mHealth solutions using blockchain technology.

Supervisory control and data acquisition (SCADA) systems are computer systems that gather and analyze real-time data, distributed control systems are specially designed automated control system that consists of geographically distributed control elements, and other smaller control systems such as programmable logic controllers are industrial solid-state computers that monitor inputs and outputs and make logic-based decisions. In recent years, there has been a lot of focus on the security of industrial control systems. Due to the advancement in information technologies, the risk of cyberattacks on industrial control system has been drastically increased. Because they are so inextricably tied to human life, any damage to them might have devastating consequences. To provide an efficient solution to such problems, this paper proposes a new approach to intrusion detection. First, the important features in the dataset are determined by the difference between the distribution of unlabeled and positive data which is deployed for the learning process. Then, a prior estimation of the class is proposed based on a support vector machine. Simulation results show that the proposed approach has better anomaly detection performance than existing algorithms.