De-anonymization Attacks on Neuroimaging Datasets
Vikram Ravindra, Ananth Grama
Abstract
Advances in imaging technologies, combined with inexpensive storage, have led to an explosion in the volume of publicly available neuroimaging datasets. Effective analyses of these images hold the potential for uncovering mechanisms that govern functioning of the human brain, and understanding various neurological diseases and disorders. The potential significance of these studies notwithstanding, a growing concern relates to the protection of privacy and confidentiality of subjects who participate in these studies. In this paper, we present a de-anonymization attack rooted in the innate uniqueness of the structure and function of the human brain. We show that the attack reveals not only the identity of an individual, but also the efficacy with which they performing cognitive tasks. Our attack relies on novel matrix analyses techniques that are used to extract discriminating features in neuroimages. These features correspond to individual-specific signatures that can be matched across datasets for highly accurate identification. We present data preprocessing, signature extraction, and matching techniques that are computationally inexpensive, and can scale to large datasets. We characterize the efficacy of our de-anonymization attacks on publicly available databases. Finally, we discuss implications of the attack and challenges associated with defending against such attacks.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Related papers
- Cracking White-box DNN Watermarks via Invariant Neuron TransformsXudong Pan, Mi Zhang, Yifan Yan, Yining Wang et al.KDD 2023 · 10 citations
- Large-scale online deanonymization with LLMsSimon Lermen, Daniel Paleka, Joshua Swanson, Michael Aerni et al.USENIX Security 2026 · 20 citations
- Effective De-identification Generative Adversarial Network for Face AnonymizationZhenzhong Kuang, Huigui Liu, Jun Yu, Aikui Tian et al.ACM MM 2021 · 43 citations
- Train to Defend: First Defense Against Cryptanalytic Neural Network Parameter Extraction AttacksAshley Kurian, Aydin AysuNeurIPS 2025 · 1 citation
- Beyond Slow Signs in High-fidelity Model ExtractionHanna Foerster, Robert Mullins, Ilia Shumailov, Jamie HayesNeurIPS 2024 · 19 citations
