ABSTRACT
The development of artificial intelligence and worldwide epidemic events has promoted the implementation of smart healthcare while bringing issues of data privacy, malicious attack, and service quality. The Medical Internet of Things (MIoT), along with the technologies of federated learning and blockchain, has become a feasible solution for these issues. In this paper, we present a blockchain-based federated learning method for smart healthcare in which the edge nodes maintain the blockchain to resist a single point of failure and MIoT devices implement the federated learning to make full of the distributed clinical data. In particular, we design an adaptive differential privacy algorithm to protect data privacy and gradient verification-based consensus protocol to detect poisoning attacks. We compare our method with two similar methods on a real-world diabetes dataset. Promising experimental results show that our method can achieve high model accuracy in acceptable running time while also showing good performance in reducing the privacy budget consumption and resisting poisoning attacks.
Subject(s)
Blockchain , Artificial Intelligence , Computer Security , Delivery of Health Care , PrivacyABSTRACT
Hepatocellular carcinoma (HCC) is one of the most frequent malignancies and a main cause of global cancer mortality. In the past decade, circular RNAs (circRNAs) have been proved to play key roles in various cancers. Previously, circ_0008450 was identified upregulated in HCC tissues by high-throughput circRNA sequencing. In this study, quantitative real-time polymerase chain reaction was used to evaluate the expression level of circ_0008450 in human HCC tumor and corresponding nontumor tissue samples, and the association between circ_0008450 expression and clinicopathologic features of patients with HCC was also analyzed. After that, the functions of circ_0008450 in biological behaviors of HCC cells were determined by cell counting kit-8, colony formation, flow cytometry, and the transwell assays. The mechanism of circ_0008450 was explored by the bioinformatic analysis and dual-luciferase reporter assay. The expression of circ_0008450 is upregulated in HCC tissue specimens and cell lines. Patients with a high circ_0008450 expression usually bear a lower 5-year survival rate. Silencing of circ_0008450 in Huh-7 cells inhibited cell viability, migration, and invasion, whereas cell apoptosis was increased. Conversely, its overexpression in HepG2 cells leads to absolutely inverse results. In addition, circ_0008450 was proved to be a sponge of miR-548p. The oncogenic role of circ_0008450 was partially attributed to its suppression on miR-548p. This study implies a new target for the treatment of HCC.
