ConfounderGAN: Protecting Image Data Privacy with Causal Confounder

Kuang, Kun; Jiang, Kelu; Liu, Furui; Wang, Zhihua; Wu, Fei

doi:10.48550/arxiv.2212.01767

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Adversarial watermarks leverage this characteristic to prevent data infringement by DNN models. Tian et al [25] create imperceptible adversarial watermarks to confound label-image mapping, to prevent private images collection from unapproved DNN on the internet. Backdoor watermarking [5,26] is used to protect the copyright of the authorized model by correlating specific categories with a trigger.…”

Section: Introductionmentioning

confidence: 99%

MedLocker: A Transferable Adversarial Watermarking for Preventing Unauthorized Analysis of Medical Image Dataset

Pu¹,

Wei²,

Song³

et al. 2023

Preprint

View full text Add to dashboard Cite

The collection of medical image datasets is a demanding and laborious process that requires significant resources. Furthermore, these medical datasets may contain personally identifiable information, necessitating measures to ensure that unauthorized access is prevented. Failure to do so could violate the intellectual property rights of the dataset owner and potentially compromise the privacy of patients. As a result, safeguarding medical datasets and preventing unauthorized usage by AI diagnostic models is a pressing challenge. To address this challenge, we propose a novel visible adversarial watermarking method for medical image copyright protection, called MedLocker. Our approach involves continuously optimizing the position and transparency of a watermark logo, which reduces the performance of the target model, leading to incorrect predictions. Importantly, we ensure that our method minimizes the impact on clinical visualization by constraining watermark positions using semantical masks (WSM), which are bounding boxes of lesion regions based on semantic segmentation. To ensure the transferability of the watermark across different models, we verify the cross-model transferability of the watermark generated on a single model. Additionally, we generate a unique watermark parameter list each time, which can be used as a certification to verify the authorization. We evaluate the performance of MedLocker on various mainstream backbones and validate the feasibility of adversarial watermarking for copyright protection on two widely-used diabetic retinopathy detection datasets. Our results demonstrate that MedLocker can effectively protect the copyright of medical datasets and prevent unauthorized users from analyzing medical images with AI diagnostic models.

show abstract

Section: Introductionmentioning

confidence: 99%