Membership Inference Attacks on Deep Regression Models for NeuroimagingDownload PDF

Published: 31 Mar 2021, Last Modified: 16 May 2023MIDL 2021Readers: Everyone
Keywords: brain age, membership inference attacks, privacy, federated-learning
TL;DR: We show realistic membership inference attacks on deep neural networks learned via either distributed or centralized training to predict brain age from MRI scans.
Abstract: Ensuring the privacy of research participants is vital, even more so in healthcare environments. Deep learning approaches to neuroimaging require large datasets, and this often necessitates sharing data between multiple sites, which is antithetical to the privacy objectives. Federated learning is a commonly proposed solution to this problem. It circumvents the need for data sharing by sharing parameters during the training process. However, we demonstrate that allowing access to parameters may leak private information even if data is never directly shared. In particular, we show that it is possible to infer if a sample was used to train the model given only access to the model prediction (black-box) or access to the model itself (white-box) and some leaked samples from the training data distribution. Such attacks are commonly referred to as \textit{Membership Inference attacks}. We show realistic Membership Inference attacks on deep learning models trained for 3D neuroimaging tasks in a centralized as well as decentralized setup. We demonstrate feasible attacks on brain age prediction models (deep learning models that predict a person's age from their brain MRI scan). We correctly identified whether an MRI scan was used in model training with a 60% to over 80% success rate depending on model complexity and security assumptions.
Registration: I acknowledge that publication of this at MIDL and in the proceedings requires at least one of the authors to register and present the work during the conference.
Authorship: I confirm that I am the author of this work and that it has not been submitted to another publication before.
Paper Type: both
Primary Subject Area: Transfer Learning and Domain Adaptation
Secondary Subject Area: Unsupervised Learning and Representation Learning
Data Set Url: UK Biobank Dataset: https://www.ukbiobank.ac.uk/
Source Latex: zip
14 Replies

Loading