Recent advances in deep learning have shown many successful stories in smart healthcare applications with data-driven insight into improving clinical institutions’ quality of care. Excellent deep learning models are heavily data-driven. The more data trained, the more robust and more generalizable the performance of the deep learning model. However, pooling the medical data into centralized storage to train a robust deep learning model faces privacy, ownership, and strict regulation challenges. Federated learning resolves the previous challenges with a shared global deep learning model using a central aggregator server. At the same time, patient data remain with the local party, maintaining data anonymity and security. In this study, first, we provide a comprehensive, up-to-date review of research employing federated learning in healthcare applications. Second, we evaluate a set of recent challenges from a data-centric perspective in federated learning, such as data partitioning characteristics, data distributions, data protection mechanisms, and benchmark datasets. Finally, we point out several potential challenges and future research directions in healthcare applications.

The advent of the fifth-generation mobile communication technology (5G) era has catalyzed significant advancements in medical diagnosis delivery, primarily driven by the surge in medical data from wearable Internet of Medical Things (IoMT) devices. Nonetheless, the IoMT paradigm grapples with challenges related to data security, privacy, constrained computational capabilities at the edge, and an inadequate architecture for handling traditionally error-prone data. In this context, our research offers: (1) an exhaustive review of large-scale medical data propelled by IoMT, (2) an exploration of the prevailing cloud-edge Artificial Intelligence (AI) framework tailored for IoMT, and (3) an insight into the application of Edge Federated Learning (EFL) in bolstering medical big data analytics to yield secure and superior diagnostic outcomes. We place a particular emphasis on the proliferation of IoMT wearable devices that incessantly stream medical data, either from patients or healthcare institutions, to centralized repositories. Furthermore, we introduce a federated cloud-edge AI blueprint designed to position computational resources proximate to the edge network, facilitating real-time diagnostic feedback to patients. We conclude by delineating prospective research trajectories in enhancing IoMT through AI integration.

Early prediction of students’ learning performance and analysis of student behavior in a virtual learning environment (VLE) are crucial to minimize the high failure rate in online courses during the COVID-19 pandemic. Nevertheless, traditional machine learning models fail to predict student performance in the early weeks due to the lack of students’ activities’ data in a week-wise timely manner (i.e., spatiotemporal feature issues). Furthermore, the imbalanced data distribution in the VLE impacts the prediction model performance. Thus, there are severe challenges in handling spatiotemporal features, imbalanced data sets, and a lack of explainability for enhancing the confidence of the prediction system. Therefore, an intelligent framework for explainable student performance prediction (ESPP) is proposed in this study in order to provide the interpretability of the prediction results. First, this framework utilized a time-series weekly student activity data set and dealt with the VLE imbalanced data distribution using a hybrid data sampling method. Then, a combination of convolutional neural network (CNN) and long short-term memory (LSTM) was employed to extract the spatiotemporal features and develop the early prediction deep learning (DL) model. Finally, the DL model was explained by visualizing and analyzing typical predictions, students’ activities’ maps, and feature importance. The numerical results of cross-validation showed that the proposed new DL model (i.e., the combined CNN-LSTM and ConvLSTM), in the early prediction cases, performed better than the baseline models of LSTM, support vector machine (SVM), and logistic regression (LR) models.

The sparse data in PM2.5 air quality monitoring systems is frequently happened on large-scale smart city sensing applications, which is collected via massive sensors. Moreover, it could be affected by inefficient node deployment, insufficient communication, and fragmented records, which is the main challenge of the high-resolution prediction system. In addition, data privacy in the existing centralized air quality prediction system cannot be ensured because the data which are mined from end sensory nodes constantly exposed to the network. Therefore, this paper proposes a novel edge computing framework, named Federated Compressed Learning (FCL), which provides efficient data generation while ensuring data privacy for PM2.5 predictions in the application of smart city sensing. The proposed scheme inherits the basic ideas of the compression technique, regional joint learning, and considers a secure data exchange. Thus, it could reduce the data quantity while preserving data privacy. This study would like to develop a green energy-based wireless sensing network system by using FCL edge computing framework. It is also one of key technologies of software and hardware co-design for reconfigurable and customized sensing devices application. Consequently, the prototypes are developed in order to validate the performances of the proposed framework. The results show that the data consumption is reduced by more than 95% with an error rate below 5%. Finally, the prediction results based on the FCL will generate slightly lower accuracy compared with centralized training. However, the data could be heavily compacted and securely transmitted in WSNs.

Advances in information technology (IT) and operation technology (OT) accelerate the development of manufacturing systems (MS) consisting of integrated circuits (ICs), modules, and systems, toward Industry 4.0. However, the existing MS does not support comprehensive identity forensics for the whole system, limiting its ability to adapt to equipment authentication difficulties. Furthermore, the development of trust imposed during their crosswise collaborations with suppliers and other manufacturers in the supply chain is poorly maintained. In this paper, a trust chain framework with a comprehensive identification mechanism is implemented for the designed MS system, which is based and created on the private blockchain in conjunction with decentralized database systems to boost the flexibility, traceability, and identification of the IC-module-system. Practical implementations are developed using a functional prototype. First, the decentralized application (DApp) and the smart contracts are proposed for constructing the new trust chain under the proposed comprehensive identification mechanism by using blockchain technology. In addition, the blockchain addresses of IC, module, and system are automatically registered to InterPlanetary File System (IPFS), individually. In addition, their corresponding hierarchical CID (content identifier) values are organized by using Merkle DAG (Directed Acyclic Graph), which is employed via the hierarchical content identifier mechanism (HCIDM) proposed in this paper. Based on insights obtained from this analysis, the trust chain based on HCIDM can be applied to any MS system, for example, this trust chain could be used to prevent the counterfeit modules and ICs employed in the monitoring system of a semiconductor factory environment. The evaluation results show that the proposed scheme could work in practice under the much lower costs, compared to the public blockchain, with a total cost of 0.002094 Ether. Finally, this research is developed an innovation trust chain mechanism that could be provided the system-level security for any MS toward Industrial 4.0 in order to meet the requirements of both manufacturing innovation and product innovation in Sustainable Development Goals (SDGs).