The Generative Pre-trained Transformer (GPT) represents a notable breakthrough in the domain of natural language processing, which is propelling us toward the development of machines that can understand and communicate using language in a manner that closely resembles that of humans. GPT is based on the transformer architecture, a deep neural network designed for natural language processing tasks. Due to their impressive performance on natural language processing tasks and ability to effectively converse, GPT have gained significant popularity among researchers and industrial communities, making them one of the most widely used and effective models in natural language processing and related fields, which motivated to conduct this review. This review provides a detailed overview of the GPT, including its architecture, working process, training procedures, enabling technologies, and its impact on various applications. In this review, we also explored the potential challenges and limitations of a GPT. Furthermore, we discuss potential solutions and future directions. Overall, this paper aims to provide a comprehensive understanding of GPT, its enabling technologies, their impact on various applications, emerging challenges, and potential solutions.
The Internet of Underwater Things (IoUT) has become widely popular in the past decade as it has huge prospects for the economy due to its applicability in various use cases such as environmental monitoring, disaster management, localization, defense, underwater exploration, and so on. However, each of these use cases poses specific challenges with respect to security, privacy, transparency, and traceability, which can be addressed by the integration of blockchain with the IoUT. Blockchain is a Distributed Ledger Technology (DLT) that consists of series of blocks chained up in chronological order in a distributed network. In this paper, we present a first-of-its-kind survey on the integration of blockchain with the IoUT. This paper initially discusses the blockchain technology and the IoUT and points out the benefits of integrating blockchain technology with IoUT systems. An overview of various applications, the respective challenges, and the possible future directions of blockchain-enabled IoUT systems is also presented in this survey, and finally, the work sheds light on the critical aspects of IoUT systems and will enable researchers to address the challenges using blockchain technology.
Diabetic Retinopathy (DR) remains a major factor contributing to vision loss worldwide, particularly among individuals with diabetes. Timely and accurate diagnosis of DR is essential to prevent vision loss and guide clinical intervention. In recent decades, Artificial Intelligence (AI) has demonstrated remarkable potential in revolutionizing DR disease identification and risk assessment. This review paper aims to perform the comprehensive analysis regarding the present status of research and developments in the field of DR using AI technologies. This comprehensive review emphasize the five categories of DR applications such as classification, detection, early diagnosis, segmentation and grading the severity of DR’s. This review article presents how AI technologies are leveraged in the current research works considering all these five DR applications. This is illustrated by using the AI-based DR workflow. It highlights the few significant case studies utilizing the AI technologies in real life problem-solving. Further, the adoption of Explainable Artificial Intelligence (XAI) techniques in the field of DR applications are discussed for model generalization and clinical validation. Finally, it presents the available benchmarking datasets, performance metrics, key challenges and future scope for DR research utilizing the AI technologies. This review work unveils several challenges and limitations with respect to input data modality, data preprocessing and data-fusion drawing the scope for further enhancements that can be made for a better DR prediction and detection.
1 June 2024 in Recent Advances in Computer Science and Communications
The Internet of Medical Things (IoMT) has emerged as a paradigm shift in healthcare, integrating the Internet of Things (IoT) with medical devices, sensors, and healthcare systems. From peripheral devices that monitor vital signs to remote patient monitoring systems and smart hospitals, IoMT provides a vast array of applications that empower healthcare professionals. However, the integration of IoMT presents numerous obstacles, such as data security, privacy concerns, interoperability, scalability, and ethical considerations. For the successful integration and deployment of IoMT, addressing these obstacles is essential. Federated Learning (FL) permits collaborative model training while maintaining data privacy in distributed environments like IoMT. By incorporating Explainable Artificial Intelligence (XAI) techniques, the resulting models become more interpretable and transparent, enabling healthcare professionals to comprehend the underlying decision-making processes. This integration not only improves the credibility of Artificial Intelligence models but also facilitates the detection of biases, errors, and peculiar patterns in the data. The combination of FL and XAI contributes to the development of more privacy-preserving, trustworthy, and explainable AI systems, which are essential for the development of dependable and ethically sound IoMT applications. Hence, the aim of this paper is to conduct a literature review on the amalgamation of FL and XAI for IoMT.
The advancement and innovations in the field of science and technology paved way for various advanced treatments in the field of medicine. They are implemented using sensors, and computer-aided designs with artificial intelligence techniques. This helps in the detection of serious health constraints at an earlier stage with appropriate treatments using decision-making techniques. One of the important health concerns that are increasing rapidly is cardiovascular disorders. This includes Arrhythmia and Myocardial Infarction. Earlier prediction and classification can protect them from serious constraints. They are diagnosed using the Electrocardiogram (ECG). To obtain accurate results, artificial intelligence techniques are implemented to extract the optimum output. The proposed system includes the detection and classification using deep learning techniques with the Internet of Things (IoT). The existing heartbeat detection system is overcome using a deep convolutional neural network. This helps in the implementation of automatic heartbeat detection and identification of abnormalities. The ECG signals are pre-processed with segmentation and feature extraction techniques. The classification and identification of constraints in the functioning of the heart are identified using optimization algorithms. The proposed system is trained, tested, and evaluated using the MIT-BIH arrhythmia database. The accuracy and efficiency of the proposed system are 99.98% using the MIT-BIH dataset.
1 July 2023 in Indonesian Journal of Electrical Engineering and Computer Science
Sensor networks have helped wireless communication systems. Over the last decade, researchers have focused on energy efficiency in wireless sensor networks. Energy-efficient routing remains unsolved. Because energyconstrained sensors have limited computing capabilities, extending their lifespan is difficult. This work offers a simple, energy-efficient data fusion technique employing zonal node information. Using the witness-based data fusion technique, the evaluated network lifetime, energy consumption, communication overhead, end-to-end delay, and data delivery ratio. Energyefficient data fusion optimizes energy utilization using squirrel search optimization and a recurrent neural network. The method allows the system to recognize a sensor with excessive energy dissipation and relocate data fusion to a more energy-efficient node. The proposed model was compared against artificial neural network-particle swarm optimization (ANN-PSO), cuckoo optimization algorithm-back propagation neural network (COABPNN), Elman neural network-whale optimization algorithm (ENN-WOA), and extreme learning machine-particle swarm optimization (ELM-PSO). The model achieved 94.50% network lifetime, 26.63% communication overhead, 93.85% data delivery ratio, 10.50 ms end-to-end delay, and 282 J energy usage.