Health Information Exchange (HIE) exhibits remarkable benefits for patient care such as improving healthcare quality and expediting coordinated care. The Office of the National Coordinator (ONC) for Health Information Technology is seeking patient-centric HIE designs that shift data ownership from providers to patients. There are multiple barriers to patient-centric HIE in the current system, such as security and privacy concerns, data inconsistency, timely access to the right records across multiple healthcare facilities. After investigating the current workflow of HIE, this paper provides a feasible solution to these challenges by utilizing the unique features of blockchain, a distributed ledger technology which is considered “unhackable”. Utilizing the smart-contract feature, which is a programmable self-executing protocol running on a blockchain, we developed a blockchain model to protect data security and patients' privacy, ensure data provenance, and provide patients full control of their health records. By personalizing data segmentation and an “allowed list” for clinicians to access their data, this design achieves patient-centric HIE. We conducted a large-scale simulation of this patient-centric HIE process and quantitatively evaluated the model's feasibility, stability, security, and robustness.

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.

Background: Data coordination across multiple health care facilities has become increasingly important for many emerging health care applications. Distrust has been recognized as a key barrier to the success of such applications. Leveraging blockchain technology could provide potential solutions tobuild trust between data providers and receivers by taking advantage of blockchain properties such as security, immutability, anonymity, decentralization, and smart contracts. Many health technologies have empirically proven that blockchain designs fit well with the needs of health care applications with certain degrees of success. However, there is a lack of robust architecture to provide a practical framework for developers to implement applications and test the performance of stability, efficiency, and scalability using standard blockchain designs. A generalized blockchain model is needed for the health care community to adopt blockchain technology and develop applications in a timely fashion. Objective: This study aimed at building a generalized blockchain architecture that provides data coordination functions, including data requests, permission granting, data exchange, and usage tracking, for a wide spectrum of health care application developments. Methods: An augmented, 3-layered blockchain architecture was built on a private blockchain network. The 3 layers, from bottom to top, are as follows: (1) incorporation of fundamental blockchain settings and smart contract design for data collection; (2) interactions between the blockchain and health care application development environment using Node.js and web3.js; and (3) a flexible development platform that supports web technologies such as HTML, https, and various programing languages. Two example applications, health information exchange (HIE) and clinical trial recruitment, were developed in our design to demonstrate the feasibility of the layered architecture. Case studies were conducted to test the performance in terms of stability, efficiency, and scalability of the blockchain system. Results: A total of 331,142 simulated HIE requests from accounts of 40,000 patients were successfully validated through this layered blockchain architecture with an average exchange time of 11.271 (SD 2.208) seconds. We also simulated a clinical trial recruitment scenario with the same set of patients and various recruitment criteria to match potential subjects using the same architecture. Potential subjects successfully received the clinical trial recruitment information and granted permission to the trial sponsors to access their health records with an average time of 3.07 seconds. Conclusions: This study proposes a generalized layered blockchain architecture that offers health technology community blockchain features for application development without requiring developers to have extensive experience with blockchain technology. The case studies tested the performance of our design and empirically proved the feasibility of the architecture in 2 relevant health application domains.

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.

Background: A clinical trial management system (CTMS) is a suite of specialized productivity tools that manage clinical trial processes from study planning to closeout. Using CTMSs has shown remarkable benefits in delivering efficient, auditable, and visualizable clinical trials. However, the current CTMS market is fragmented, and most CTMSs fail to meet expectations because of their inability to support key functions, such as inconsistencies in data captured across multiple sites. Blockchain technology, an emerging distributed ledger technology, is considered to potentially provide a holistic solution to current CTMS challenges by using its unique features, such as transparency, traceability, immutability, and security. Objective: This study aimed to re-engineer the traditional CTMS by leveraging the unique properties of blockchain technology to create a secure, auditable, efficient, and generalizable CTMS. Methods: A comprehensive, blockchain-based CTMS that spans all stages of clinical trials, including a sharable trial master file system; a fast recruitment and simplified enrollment system; a timely, secure, and consistent electronic data capture system; a reproducible data analytics system; and an efficient, traceable payment and reimbursement system, was designed and implemented using the Quorum blockchain. Compared with traditional blockchain technologies, such as Ethereum, Quorum blockchain offers higher transaction throughput and lowers transaction latency. Case studies on each application of the CTMS were conducted to assess the feasibility, scalability, stability, and efficiency of the proposed blockchain-based CTMS. Results: A total of 21.6 million electronic data capture transactions were generated and successfully processed through blockchain, with an average of 335.4 transactions per second. Of the 6000 patients, 1145 were matched in 1.39 seconds using 10 recruitment criteria with an automated matching mechanism implemented by the smart contract. Key features, such as immutability, traceability, and stability, were also tested and empirically proven through case studies. Conclusions: This study proposed a comprehensive blockchain-based CTMS that covers all stages of the clinical trial process. Compared with our previous research, the proposed system showed an overall better performance. Our system design, implementation, and case studies demonstrated the potential of blockchain technology as a potential solution to CTMS challenges and its ability to perform more health care tasks.

Background: Earlier studies have indicated a relatively higher risk of occurring meningioma among female breast cancer survivors and have suggested that tamoxifen might decrease this risk. The present study evaluated whether tamoxifen use in breast cancer patients can reduce the risk of meningioma. Methods: We designed a population-based cohort study by using data from the National Health Insurance system of Taiwan to assess this issue. Between January 1, 2000 and December 31, 2008, women with breast cancer and of age ≥20 years were included. They were divided into 2 groups: those who had not received tamoxifen therapy and those who had. The Cox’s proportion hazard regression analysis was conducted to estimate the effects of tamoxifen treatment and the subsequent meningioma risk. Results: We identified a total of 50,442 tamoxifen users and 30,929 non-tamoxifen users. Tamoxifen users had a borderline significantly lower overall risk of meningioma than non-tamoxifen users [adjusted hazard ratio (aHR) = 0.64, 95% confidence interval (95% CI) = 0.40–1.02]. A statistically significant difference was found in those patients with tamoxifen treatment duration longer than 1500 days (aHR = 0.42, 95% CI = 0.19–0.91) or with cumulative dosage exceeding 26320 mg (aHR = 0.44, 95% CI = 0.22–0.88). Furthermore, no statistically significant joint effect of aromatase inhibitors and tamoxifen on the occurrence of meningioma among breast cancer patients was seen. Conclusion: Tamoxifen users had a non-significantly (36%) lower risk of developing meningioma than did tamoxifen non-users; however, our data indicated that tamoxifen therapy is associated with a reduced meningioma risk for Taiwanese breast cancer patients receiving long duration or high cumulative dosage treatment with tamoxifen.