Editorial: Protecting privacy in neuroimaging analysis: balancing data sharing and privacy preservation

Frontiers in Neuroinformatics, Journal Year: 2025, Volume and Issue: 18

Published: Jan. 7, 2025

Neuroimaging is an indispensable tool in neuroscience and medical research, enabling precise investigations into brain structure function (Yen, Lin, Chiang 2023;Yan et al. 2022;Shoeibi 2023;Botvinik-Nezer Wager 2023;Leite 2024;Wager Smith 2003). Techniques such as Magnetic Resonance Imaging (MRI) generate vast amounts of sensitive data, rich insights yet fraught with privacy challenges (Saponaro 2022;Cali 2023;Li 2020;Zou 2024;Acar 2023). As scientific progress depends on data sharing collaboration (Martone 2023), balancing these needs robust preservation has become a critical concern (Zhang 2020). This special issue addresses this challenge by exploring innovative methodologies, frameworks, technologies that advance the field while safeguarding individual privacy.The aims to promote interdisciplinary research privacy-preserving solutions for neuroimaging analysis, ensuring compliance ethical legal standards (Li It seeks balance utility protections fostering methods anonymization, leveraging AI tools federated learning differential privacy, aligning global governance frameworks (Zou 2024;Jeon 2020;Dwork 2006;Abadi 2016). serves roadmap platform dialogue among neuroscientists, researchers, ethicists, policymakers.This features five papers exemplify breadth depth at intersection neuroimaging, artificial intelligence. Each contribution highlights unique facet landscape, collectively offering comprehensive exploration field's current state future potential.The first paper 1 tackles pervasive inflated effect sizes small-sample studies, undermines reproducibility generalizability. By employing hierarchical Bayesian models, authors demonstrate how statistical recalibration can improve reliability findings collaborative metaanalyses across studies. methodological sets foundation shared not only secure but also statistically robust.The second 2 explores AI-driven segmentation intracranial haemorrhage detection CT scans. Leveraging self-supervised weakly-supervised learning, study need label-efficient minimize reliance large, annotated datasets. work showcases innovations enhance efficiency maintain particularly resource-constrained environments where annotation bottleneck.Federated takes centre stage third fourth papers, both which highlight its potential decentralized analysis. The 3 introduces framework Alzheimer's disease detection, incorporating aggregation techniques protect during model training. Similarly, 4 presents Sparse Federated Learning (NeuroSFL), optimizes communication focusing sparse sub-networks. Together, studies underscore adaptability scalability cornerstone privacypreserving research.The final 5 adopts broader lens, examining alignment neuroinformatics practices. identifying gaps existing regulations proposing strategies harmonization, provide integrating within complex landscape governance. emphasizes importance technical advancements principles, trust transparency research.Artificial intelligence driving force behind many contributions issue, powerful privacy. methodologies explainable enable analysis trustworthiness (Yuste 2023;White, Blok, Calhoun 2022;Yang 2022). These address challenges, mitigating risks remains private without compromising utility. particular, emerges transformative approach, allowing researchers train models collaboratively raw data. sparsity-focused presented scale meet demands heterogeneous Complementary blockchain hold promise further enhancing security accountability, though their integration routine workflows challenge.Despite advancements, significant persist. Balancing fundamental tension, often introduce trade-offs performance or 2023;Mitrovska 2024). For instance, are susceptible degradation non-IID (non-independent identically distributed) settings, common scenario neuroimaging. computational may limit accessibility smaller institutions, exacerbating inequities field.Ethical societal add another layer complexity (Aboy, Minssen, Vayena 2024;van Kolfschooten van Oirschot Cognitive informed consent, equitable access benefits ongoing concerns (Bublitz, Molnár-Gábor, Soekadar rapid evolution outpaces development regulatory creating misalignments between technological capabilities oversight (Ratto Trabucco 2023;Ienca Ignatiadis 2020;Wajnerman Paz 2022;Jwa Martinez-Martin 2024;Yuste 2017;Genser, Damianos, Yuste Addressing will require sustained stakeholders, including technologists, policymakers (Ligthart 2023;Bublitz, 2024).This emphasizing advancing maintaining presenting cutting-edge practical real-world applications, offer research. works coexist, disciplines.The ubiquity amplifies dynamic adaptive evolve alongside advancements. intersects innovation, static siloed insufficient overlapping Instead, flexible approaches align AI's essential responsible progress.This testament tackling technology, ethics, neuroscience. it lays groundwork thrives environment trust, transparency, progress. We invite readers engage contributions, conversation shaping more beyond. would have been possible dedication authors, whose forms foundation, reviewers, constructive feedback ensured rigor, communities, drive discovery.

Language: Английский

---

A comprehensive review of neurotransmitter modulation via artificial intelligence: A new frontier in personalized neurobiochemistry

Computers in Biology and Medicine, Journal Year: 2025, Volume and Issue: 189, P. 109984 - 109984

Published: March 14, 2025

Language: Английский

---

A Community Effort to Develop Common Data Elements for Pre-Clinical Spinal Cord Injury Research

Neurotrauma Reports, Journal Year: 2025, Volume and Issue: 6(1), P. 391 - 401

Published: Jan. 1, 2025

Language: Английский

---

Interdisciplinary and Collaborative Training in Neuroscience: Insights from the Human Brain Project Education Programme

Neuroinformatics, Journal Year: 2024, Volume and Issue: 22(4), P. 657 - 678

Published: Nov. 6, 2024

Abstract Neuroscience education is challenged by rapidly evolving technology and the development of interdisciplinary approaches for brain research. The Human Brain Project (HBP) Education Programme aimed to address need expertise in research equipping a new generation researchers with skills across neuroscience, medicine, information technology. Over its ten year duration, programme engaged over 1,300 experts attracted more than 5,500 participants from various scientific disciplines blended learning curriculum, specialised schools workshops, events fostering dialogue among early-career researchers. Key principles programme’s approach included interdisciplinarity, adaptability landscape infrastructure, collaborative environment focus on empowering Following conclusion, we provide here an analysis in-depth view diverse range educational formats events. Our results show that achieved success wide geographic reach, diversity participants, establishment transversal collaborations. Building these experiences achievements, describe how leveraging digital tools platforms provides accessible highly training, which can enhance existing programmes next working decentralised European spaces. Finally, present lessons learnt so similar initiatives may improve upon our experience incorporate suggestions into their own programme.

Language: Английский

---

Connecting the dots in neuroscience research: The future of evidence synthesis

Experimental Neurology, Journal Year: 2024, Volume and Issue: 384, P. 115047 - 115047

Published: Nov. 5, 2024

Language: Английский

---