Project description:Medical healthcare centers are envisioned as a promising paradigm to handle the massive volume of data for COVID-19 patients using artificial intelligence (AI). Traditionally, AI techniques require centralized data collection and training models within a single organization. This practice can be considered a weakness as it leads to several privacy and security concerns related to raw data communication. To overcome this weakness and secure raw data communication, we propose a blockchain-based federated learning framework that provides a solution for collaborative data training. The proposed framework enables the coordination of multiple hospitals to train and share encrypted federated models while preserving data privacy. Blockchain ledger technology provides decentralization of federated learning models without relying on a central server. Moreover, the proposed homomorphic encryption scheme encrypts and decrypts the gradients of the model to preserve privacy. More precisely, the proposed framework: (i) train the local model by a novel capsule network for segmentation and classification of COVID-19 images, (ii) furthermore, we use the homomorphic encryption scheme to secure the local model that encrypts and decrypts the gradients, (iii) finally, the model is shared over a decentralized platform through the proposed blockchain-based federated learning algorithm. The integration of blockchain and federated learning leads to a new paradigm for medical image data sharing over the decentralized network. To validate our proposed model, we conducted comprehensive experiments and the results demonstrate the superior performance of the proposed scheme.
Project description:Low-Power Wide-Area Network (LPWAN) is one of the enabling technologies of the Internet of Things (IoT), and focuses on providing long distance connectivity for a vast amount of smart devices. Currently, LoRa is one of the leading LPWAN solutions available for public use. In LPWANs, especially in LoRa, security is a major concern due to the resource constraints of the devices, the sensitivity level of the transmitted data, the large amount of connected devices, among other reasons. This paper studies the key management mechanism of LoRaWAN environments. A secure architecture for key management based on smart contracts and permissioned blockchain to enhance security and availability in LoRaWAN networks is proposed. To demonstrate the feasibility of the proposed blockchain-based LoRaWAN architecture, a working prototype has been created using open-source tools and commodity hardware. Performance analysis shows that the prototype presents similar execution time and latency values, when compared to a traditional system, especially for small and medium-sized LoRaWAN networks. We also discuss why the proposed solution can be used in environments with a large number of end-devices.
Project description:The successful implementation of pharmacogenetics (PGx) into clinical practice requires patient genomic data to be shared between stakeholders in multiple settings. This creates a number of barriers to widespread adoption of PGx, including privacy concerns related to the storage and movement of identifiable genomic data. Informatic solutions that support secure and equitable data access for genomic data are therefore important to PGx. Here we propose a methodology that uses smart contracts implemented on a blockchain-based framework, PGxChain, to address this issue. The design requirements for PGxChain were identified through a systematic literature review, identifying technical challenges and barriers impeding the clinical implementation of pharmacogenomics. These requirements included security and privacy, accessibility, interoperability, traceability and legal compliance. A proof-of-concept implementation based on Ethereum was then developed that met the design requirements. PGxChain's performance was examined using Hyperledger Caliper for latency, throughput, and transaction success rate. The findings clearly indicate that blockchain technology offers considerable potential to advance pharmacogenetic data sharing, particularly with regard to PGx data security and privacy, large-scale accessibility of PGx data, PGx data interoperability between multiple health care providers and compliance with data-sharing laws and regulations.
Project description:DataCryptChain is an open-source standard combining blockchain with advanced encryption ensuring research data remains private, secure, shareable, and tamper-proof. Ability to detect intentional tampering of data was measured, and user experience was evaluated. In this study, simulated datasets were randomized to be tampered with or not tampered with, and detection of tampering was measured. Volunteer's ability to complete assigned tasks using the software was evaluated. Among 10000 simulated datasets (4436 randomized to tampering) there was 100% sensitivity and specificity for detection. All volunteers successfully installed DataCryptChain and 5/6 completed their tasks. All participants were able to transmit data without ever exposing unencrypted data and with no need to share passwords. Several deficiencies in the user experience were noted. Importantly, the test users felt that although they would be willing to use DataCryptChain in practice, it would need a more user-friendly interface. This study demonstrates a novel algorithm using blockchain and asymmetric encryption that, although previously documented theoretically, has never been published as a working software package. While DataCryptChain has 100% sensitivity and specificity for detecting data tampering, further development is needed to improve the user experience.
Project description:BACKGROUND:The potential of blockchain technology to achieve strategic goals, such as value-based care, is increasingly being recognized by both researchers and practitioners. However, current research and practices lack comprehensive approaches for evaluating the benefits of blockchain applications. OBJECTIVE:The goal of this study was to develop a framework for holistically assessing the performance of blockchain initiatives in providing value-based care by extending the existing balanced scorecard (BSC) evaluation framework. METHODS:Based on a review of the literature on value-based health care, blockchain technology, and methods for evaluating initiatives in disruptive technologies, we propose an extended BSC method for holistically evaluating blockchain applications in the provision of value-based health care. The proposed method extends the BSC framework, which has been extensively used to measure both financial and nonfinancial performance of organizations. The usefulness of our proposed framework is further demonstrated via a case study. RESULTS:We describe the extended BSC framework, which includes five perspectives (both financial and nonfinancial) from which to assess the appropriateness and performance of blockchain initiatives in the health care domain. CONCLUSIONS:The proposed framework moves us toward a holistic evaluation of both the financial and nonfinancial benefits of blockchain initiatives in the context of value-based care and its provision.
Project description:Federated learning (FL) enables users to train the global model cooperatively without exposing their private data across the engaged parties, which is widely used in privacy-sensitive business. However, during the life cycle of FL models, both adversaries' attacks and ownership generalization threaten the FL models' copyright and affect the models' reliability. To address these problems, existing model watermarking techniques can be used to verify FL model's ownership. However, due to the lack of credible binding from "model extracted watermarks" to "ownership verification", it is difficult to form a closed-loop watermarking framework for copyright protection. Therefore, starting from the shortcomings of the current watermark verification scheme, this article proposed WFB, a blockchain-empowered watermarking framework for ownership verification of federated models. Firstly, we propose a improved watermark generation algorithm to solve the credibility issue of watermarks. Secondly, we propose a watermark embedding method in federated learning, while blockchain technology is used to ensure the credible storage of watermark information throughout the process. Thirdly, the credibility of ownership verification is improved because of the watermark authenticity. Experimental results demonstrate the fidelity, effectiveness and robustness of WFB, with other superiorities such as improving process security and traceability.
Project description:Presently modern technology makes a significant contribution to the transition from traditional healthcare to smart healthcare systems. Mobile health (mHealth) uses advances in wearable sensors, telecommunications and the Internet of Things (IoT) to propose a new healthcare concept centered on the patient. Patients' real-time remote continuous health monitoring, remote diagnosis, treatment, and therapy is possible in an mHealth system. However, major limitations include the transparency, security, and privacy of health data. One possible solution to this is the use of blockchain technologies, which have found numerous applications in the healthcare domain mainly due to theirs features such as decentralization (no central authority is needed), immutability, traceability, and transparency. We propose an mHealth system that uses a private blockchain based on the Ethereum platform, where wearable sensors can communicate with a smart device (a smartphone or smart tablet) that uses a peer-to-peer hypermedia protocol, the InterPlanetary File System (IPFS), for the distributed storage of health-related data. Smart contracts are used to create data queries, to access patient data by healthcare providers, to record diagnostic, treatment, and therapy, and to send alerts to patients and medical professionals.
Project description:In the last decade, Elliptic Curves (ECs) have shown their efficacy as a safe fundamental component in encryption systems, mainly when used in Pseudorandom Number Generator (PRNG) design. This paper proposes a framework for designing EC-based PRNG and maps recent PRNG design techniques into the framework, classifying them as iterative and non-iterative. Furthermore, a PRNG is designed based on the framework and verified using the National Institute of Standards and Technology (NIST) statistical test suite. The PRNG is then utilized in an image encryption system where statistical measures, differential attack measures, the NIST statistical test suite, and system key sensitivity analysis are used to demonstrate the system's security. The results are good and promising as compared with other related work.
Project description:In the medical system, the verification, preservation and synchronization of electronic medical records has always been a difficult problem, and the random dissemination of patient records will bring various risks to patient privacy. Therefore, how to achieve secure data sharing on the basis of ensuring users' personal privacy becomes the key. In recent years, blockchain has been proposed to be a promising solution to achieve data sharing with security and privacy preservation due to its advantages of immutability. So, a distributed electronic medical records searchable scheme was proposed by leveraging blockchain and smart contract technology. Firstly, we perform a hash calculation on the electronic medical data and store the corresponding value on the blockchain to ensure its integrity and authenticity. Then, we encrypt the electronic medical data and store it in the interplanetary file system which is a distributed storage protocol. These operations not only can solve centralized data store of servers of several medical institutions, but also be good at lowering stress from data store and high-frequency access to blockchain. Secondly, the encrypted keyword index information of electronic medical records was stored on the Ethereum blockchain, meanwhile a smart contract deployed in the Ethereum blockchain is used to realize keyword search instead of depending on a centralized third party. Furthermore, we use attribute-based encryption scheme to ensure that only the attributes meeting the access policy can decrypt the encrypted electronic medical records. Finally, our performance analysis and security analysis show that the scheme is secure and efficient.