Sequre: a high-performance framework for secure multiparty computation enables biomedical data sharing DOI Creative Commons
Haris Smajlović, Ariya Shajii, Bonnie Berger

et al.

Genome biology, Journal Year: 2023, Volume and Issue: 24(1)

Published: Jan. 11, 2023

Abstract Secure multiparty computation (MPC) is a cryptographic tool that allows on top of sensitive biomedical data without revealing private information to the involved entities. Here, we introduce Sequre, an easy-to-use, high-performance framework for developing performant MPC applications. Sequre offers set automatic compile-time optimizations significantly improve performance applications and incorporates syntax Python programming language facilitate rapid application development. We demonstrate its usability various bioinformatics tasks showing up 3–4 times increased speed over existing pipelines with 7-fold reductions in codebase sizes.

Language: Английский

GA4GH: International policies and standards for data sharing across genomic research and healthcare DOI Creative Commons
Heidi L. Rehm, Angela Page, Lindsay Smith

et al.

Cell Genomics, Journal Year: 2021, Volume and Issue: 1(2), P. 100029 - 100029

Published: Nov. 1, 2021

The Global Alliance for Genomics and Health (GA4GH) aims to accelerate biomedical advances by enabling the responsible sharing of clinical genomic data through both harmonized aggregation federated approaches. decreasing cost sequencing (along with other genome-wide molecular assays) increasing evidence its utility will soon drive generation sequence from tens millions humans, levels diversity. In this perspective, we present GA4GH strategies addressing major challenges revolution. We describe organization, which is fueled development efforts eight Work Streams informed needs 24 Driver Projects key stakeholders. suite secure, interoperable technical standards policy frameworks review current status standards, their relevance domains research care, future plans GA4GH. Broad international participation in building, adopting, deploying catalyze an unprecedented effort that be critical advancing medicine ensuring all populations can access benefits.

Language: Английский

Citations

186
Heterogeneous Federated Learning: State-of-the-art and Research Challenges DOI Open Access
Mang Ye, Xiuwen Fang, Bo Du

et al.

ACM Computing Surveys, Journal Year: 2023, Volume and Issue: 56(3), P. 1 - 44

Published: Sept. 27, 2023

Federated learning (FL) has drawn increasing attention owing to its potential use in large-scale industrial applications. Existing FL works mainly focus on model homogeneous settings. However, practical typically faces the heterogeneity of data distributions, architectures, network environments, and hardware devices among participant clients. Heterogeneous Learning (HFL) is much more challenging, corresponding solutions are diverse complex. Therefore, a systematic survey this topic about research challenges state-of-the-art essential. In survey, we firstly summarize various HFL from five aspects: statistical heterogeneity, communication device additional challenges. addition, recent advances reviewed new taxonomy existing methods proposed with an in-depth analysis their pros cons. We classify three different levels according procedure: data-level, model-level, server-level. Finally, several critical promising future directions discussed, which may facilitate further developments field. A periodically updated collection available at https://github.com/marswhu/HFL_Survey.

Language: Английский

Citations

147
A systematic review of homomorphic encryption and its contributions in healthcare industry DOI Creative Commons
Kundan Munjal, Rekha Bhatia

Complex & Intelligent Systems, Journal Year: 2022, Volume and Issue: 9(4), P. 3759 - 3786

Published: May 3, 2022

Abstract Cloud computing and cloud storage have contributed to a big shift in data processing its use. Availability accessibility of resources with the reduction substantial work is one main reasons for revolution. With this revolution, outsourcing applications are great demand. The client uses service by uploading their finally gets result it. It benefits users greatly, but it also exposes sensitive third-party providers. In healthcare industry, patient health records digital patient’s medical history kept hospitals or care Patient stored centers processing. Before doing computations on data, traditional encryption techniques decrypt original form. As result, information lost. Homomorphic can protect allowing be processed an encrypted form such that only accessible paper, attempt made present systematic review homomorphic cryptosystems categorization evolution over time. addition, paper includes cryptosystem contributions healthcare.

Language: Английский

Citations

122
Sociotechnical safeguards for genomic data privacy DOI Open Access
Zhiyu Wan, James Hazel, Ellen Wright Clayton

et al.

Nature Reviews Genetics, Journal Year: 2022, Volume and Issue: 23(7), P. 429 - 445

Published: March 4, 2022

Language: Английский

Citations

90
Decentralised clinical trials: ethical opportunities and challenges DOI Creative Commons
Effy Vayena, Alessandro Blasimme, Jeremy Sugarman

et al.

The Lancet Digital Health, Journal Year: 2023, Volume and Issue: 5(6), P. e390 - e394

Published: April 25, 2023

Language: Английский

Citations

56
Metaverse Wearables for Immersive Digital Healthcare: A Review DOI Creative Commons
Kisoo Kim,

Hyosill Yang,

Ji-Hun Lee

et al.

Advanced Science, Journal Year: 2023, Volume and Issue: 10(31)

Published: Sept. 22, 2023

The recent exponential growth of metaverse technology has been instrumental in reshaping a myriad sectors, not least digital healthcare. This comprehensive review critically examines the landscape and future applications wearables toward immersive key technologies advancements that have spearheaded metamorphosis are categorized, encapsulating all-encompassed extended reality, such as virtual augmented mixed other haptic feedback systems. Moreover, fundamentals their deployment assistive healthcare (especially for rehabilitation), medical nursing education, remote patient management treatment investigated. potential benefits integrating into paradigms multifold, encompassing improved prognosis, enhanced accessibility to high-quality care, high standards practitioner instruction. Nevertheless, these without inherent challenges untapped opportunities, which span privacy protection, data safeguarding, innovation artificial intelligence. In summary, research trajectories circumvent hurdles also discussed, further augmenting incorporation within infrastructures post-pandemic era.

Language: Английский

Citations

47
Decentralized Federated Learning: A Survey and Perspective DOI
Liangqi Yuan, Ziran Wang, Lichao Sun

et al.

IEEE Internet of Things Journal, Journal Year: 2024, Volume and Issue: 11(21), P. 34617 - 34638

Published: May 30, 2024

Federated learning (FL) has been gaining attention for its ability to share knowledge while maintaining user data, protecting privacy, increasing efficiency, and reducing communication overhead. Decentralized FL (DFL) is a decentralized network architecture that eliminates the need central server in contrast centralized (CFL). DFL enables direct between clients, resulting significant savings resources. In this paper, comprehensive survey profound perspective are provided DFL. First, review of methodology, challenges, variants CFL conducted, laying background Then, systematic detailed on introduced, including iteration order, protocols, topologies, paradigm proposals, temporal variability. Next, based definition DFL, several extended categorizations proposed with state-of-the-art (SOTA) technologies. Lastly, addition summarizing current challenges some possible solutions future research directions also discussed.

Language: Английский

Citations

44
PPML-Omics: A privacy-preserving federated machine learning method protects patients’ privacy in omic data DOI Creative Commons
Juexiao Zhou, Siyuan Chen, Yulian Wu

et al.

Science Advances, Journal Year: 2024, Volume and Issue: 10(5)

Published: Jan. 31, 2024

Modern machine learning models toward various tasks with omic data analysis give rise to threats of privacy leakage patients involved in those datasets. Here, we proposed a secure and privacy-preserving method (PPML-Omics) by designing decentralized differential private federated algorithm. We applied PPML-Omics analyze from three sequencing technologies addressed the concern major under representative deep models. examined breaches depth through attack experiments demonstrated that could protect patients' privacy. In each these applications, was able outperform methods comparison same level guarantee, demonstrating versatility simultaneously balancing capability utility analysis. Furthermore, gave theoretical proof PPML-Omics, suggesting first mathematically guaranteed robust generalizable empirical performance protecting data.

Language: Английский

Citations

19
A Review of Blockchain-Based Secure Sharing of Healthcare Data DOI Creative Commons
Xi Peng, Xinglong Zhang, Lian Wang

et al.

Applied Sciences, Journal Year: 2022, Volume and Issue: 12(15), P. 7912 - 7912

Published: Aug. 7, 2022

Medical data contains multiple records of patient that are important for subsequent treatment and future research. However, it needs to be stored shared securely protect the privacy data. Blockchain is widely used in management healthcare because its decentralized tamper-proof features. In order study development blockchain healthcare, this paper evaluates from various perspectives. We analyze blockchain-based approaches different application scenarios. These electronic medical record sharing, Internet Things federal learning. The results show smart contracts have a natural advantage field since they traceable. Finally, challenges directions discussed, which can help drive forward.

Language: Английский

Citations

64
Lead federated neuromorphic learning for wireless edge artificial intelligence DOI Creative Commons
Helin Yang, Kwok‐Yan Lam, Liang Xiao

et al.

Nature Communications, Journal Year: 2022, Volume and Issue: 13(1)

Published: July 25, 2022

In order to realize the full potential of wireless edge artificial intelligence (AI), very large and diverse datasets will often be required for energy-demanding model training on resource-constrained devices. This paper proposes a lead federated neuromorphic learning (LFNL) technique, which is decentralized energy-efficient brain-inspired computing method based spiking neural networks. The proposed technique enable devices exploit brain-like biophysiological structure collaboratively train global while helping preserve privacy. Experimental results show that, under situation uneven dataset distribution among devices, LFNL achieves comparable recognition accuracy existing AI techniques, substantially reducing data traffic by >3.5× computational latency >2.0×. Furthermore, significantly reduces energy consumption >4.5× compared standard with slight loss up 1.5%. Therefore, can facilitate development AI.

Language: Английский

Citations

61