Explainable Federated Medical Image Analysis Through Causal Learning and Blockchain.

Federated learning (FL) enables collaborative training of machine learning models across distributed medical data sources without compromising privacy. However, applying FL to medical image analysis presents challenges like high communication overhead and data heterogeneity. This paper proposes novel FL techniques using explainable artificial intelligence (XAI) for efficient, accurate, and trustworthy analysis. A heterogeneity-aware causal learning approach selectively sparsifies model weights based on their causal contributions, significantly reducing communication requirements while retaining performance and improving interpretability. Furthermore, blockchain provides decentralized quality assessment of client datasets. The assessment scores adjust aggregation weights so higher-quality data has more influence during training, improving model generalization. Comprehensive experiments show our XAI-integrated FL framework enhances efficiency, accuracy and interpretability. The causal learning method decreases communication overhead while maintaining segmentation accuracy. The blockchain-based data valuation mitigates issues from low-quality local datasets. Our framework provides essential model explanations and trust mechanisms, making FL viable for clinical adoption in medical image analysis.

Towards Large-Scale and Privacy-Preserving Contact Tracing in COVID-19 Pandemic: A Blockchain Perspective.

Activity-tracking applications and location-based services using short-range communication (SRC) techniques have been abruptly demanded in the COVID-19 pandemic, especially for automated contact tracing. The attention from both public and policy keeps raising on related practical problems, including 1) how to protect data security and location privacy? 2) how to efficiently and dynamically deploy SRC Internet of Thing (IoT) witnesses to monitor large areas? To answer these questions, in this paper, we propose a decentralized and permissionless blockchain protocol, named Bychain. Specifically, 1) a privacy-preserving SRC protocol for activity-tracking and corresponding generalized block structure is developed, by connecting an interactive zero-knowledge proof protocol and the key escrow mechanism. As a result, connections between personal identity and the ownership of on-chain location information are decoupled. Meanwhile, the owner of the on-chain location data can still claim its ownership without revealing the private key to anyone else. 2) An artificial potential field-based incentive allocation mechanism is proposed to incentivize IoT witnesses to pursue the maximum monitoring coverage deployment. We implemented and evaluated the proposed blockchain protocol in the real-world using the Bluetooth 5.0. The storage, CPU utilization, power consumption, time delay, and security of each procedure and performance of activities are analyzed. The experiment and security analysis is shown to provide a real-world performance evaluation.

Blockchain localization spoofing detection based on fuzzy AHP in IoT systems.

Location spoof detection is a major component of location proofing mechanisms in internet of things (IoT), and it is significant for the system to assess the trustworthiness of the location data associated with the user. Unlike the work that employs physical layer features, we interest in building the infrastructure for a solution to establish location spoofing detection capabilities in blockchain-based IoT systems. In detail, at the node and the mobile trajectory level, we create an IoT system for evaluating the trustworthiness of location proofs with blockchain location system features. A blockchain-based multilayer fuzzy hierarchical analysis process (AHP) evaluation method is contemplated to detect location spoofing in the IoT system. Simulation results indicate the proposed method has a superior performance and provides a basis for the trustworthiness assessment of location proofs.

Discovery of RG7834: The First-in-Class Selective and Orally Available Small Molecule Hepatitis B Virus Expression Inhibitor with Novel Mechanism of Action.

Chronic hepatitis B virus (HBV) infection is a serious public health burden, and current therapies cannot achieve satisfactory cure rate. There are high unmet medical needs of novel therapeutic agents with differentiated mechanism of action (MOA) from the current standard of care. RG7834, a compound from the dihydroquinolizinone (DHQ) chemical series, is a first-in-class highly selective and orally bioavailable HBV inhibitor which can reduce both viral antigens and viral DNA with a novel mechanism of action. Here we report the discovery of RG7834 from a phenotypic screening and the structure-activity relationship (SAR) of the DHQ chemical series. RG7834 can selectively inhibit HBV but not other DNA or RNA viruses in a virus panel screening. Both in vitro and in vivo profiles of RG7834 are described herein, and the data support further development of this compound as a chronic HBV therapy.

Pilot and pivotal study to evaluate the bioequivalence of two paroxetine 40 mg tablet formulations in healthy Chinese subjects.

The purpose of this study was to conduct a pilot study in order to obtain reliable results for further planning of a well-designed pivotal trial comparing the bioequivalence (BE) of two paroxetine tablet formulations in healthy Chinese subjects. Before conducting the pivotal trial, the pilot trial enrolled 14 subjects to help in study design, establishing the recruitment period, determining pharmacokinetics (PK) time points and sample size, and assessing BE of the two formulations. The single-center, randomized, open-label, single-dose, two period crossover study with a 7-day washout interval was conducted after obtaining information from the fasted pilot trial in 72 healthy volunteers for a pivotal study under fed and fasted conditions, respectively. There were 19 PK sample collection time points employed in both the pilot and pivotal trials. A sensitive and specific liquid chromatography- tandem mass spectrometry (LC-MS/ MS) method was developed and validated for determining paroxetine in human plasma. BE between two articles was determined by calculating 90% confidence intervals (CIs) for the ratio of Cmax 91.38 - 110.39% for the pilot trial, 99.81 - 114.08% for pivotal trial under fasted condition, and 94.06 - 110.41% for pivotal trial under fed condition, AUC(0-t) 96.06 - 110.52% for pilot study, 100.88 - 113.05% for the pivotal trial under fasted condition, and 97.08 - 106.06% for pivotal study under fed condition, and AUC(0-∞) 96.17 - 110.42% for the pilot study, 100.85 - 112.81% for the pivotal trial under fasted condition and 97.22 - 106.14% for the pivotal study under fed condition, respectively. These values for the test and reference products are within the 80 - 125% interval proposed by FDA and EMEA. It was concluded that the proposed method was successfully applied to a PK study in healthy Chinese volunteers, and results showed from both the pilot and pivotal studies that the two paroxetine formulations are bioequivalent in their rates and extent of absorption.