ENSEI: Efficient Secure Inference via Frequency-Domain Homomorphic Convolution for Privacy-Preserving Visual Recognition

Bian, Song; Wang, Tianchen; Hiromoto, Masayuki; Shi, Yiyu; Satō, Takashi

doi:10.1109/cvpr42600.2020.00942

Cited by 28 publications

(14 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, the conventional 𝑘-nearest neighbor algorithm was combined with DP [ 232 ] for privacy-preserving face attribute recognition and person reidentification. Homomorphic convolution was proposed by combining HE and secret sharing [ 233 ] for visual object detection, and adversarial perturbation was devised to prevent disclosure of biometric information in finger-selfie images [ 234 ].…”

Section: Resultsmentioning

confidence: 99%

Data Anonymization for Pervasive Health Care: Systematic Literature Mapping Study

Zuo¹,

Watson²,

Budgen³

et al. 2021

JMIR Med Inform

View full text Add to dashboard Cite

Background Data science offers an unparalleled opportunity to identify new insights into many aspects of human life with recent advances in health care. Using data science in digital health raises significant challenges regarding data privacy, transparency, and trustworthiness. Recent regulations enforce the need for a clear legal basis for collecting, processing, and sharing data, for example, the European Union’s General Data Protection Regulation (2016) and the United Kingdom’s Data Protection Act (2018). For health care providers, legal use of the electronic health record (EHR) is permitted only in clinical care cases. Any other use of the data requires thoughtful considerations of the legal context and direct patient consent. Identifiable personal and sensitive information must be sufficiently anonymized. Raw data are commonly anonymized to be used for research purposes, with risk assessment for reidentification and utility. Although health care organizations have internal policies defined for information governance, there is a significant lack of practical tools and intuitive guidance about the use of data for research and modeling. Off-the-shelf data anonymization tools are developed frequently, but privacy-related functionalities are often incomparable with regard to use in different problem domains. In addition, tools to support measuring the risk of the anonymized data with regard to reidentification against the usefulness of the data exist, but there are question marks over their efficacy. Objective In this systematic literature mapping study, we aim to alleviate the aforementioned issues by reviewing the landscape of data anonymization for digital health care. Methods We used Google Scholar, Web of Science, Elsevier Scopus, and PubMed to retrieve academic studies published in English up to June 2020. Noteworthy gray literature was also used to initialize the search. We focused on review questions covering 5 bottom-up aspects: basic anonymization operations, privacy models, reidentification risk and usability metrics, off-the-shelf anonymization tools, and the lawful basis for EHR data anonymization. Results We identified 239 eligible studies, of which 60 were chosen for general background information; 16 were selected for 7 basic anonymization operations; 104 covered 72 conventional and machine learning–based privacy models; four and 19 papers included seven and 15 metrics, respectively, for measuring the reidentification risk and degree of usability; and 36 explored 20 data anonymization software tools. In addition, we also evaluated the practical feasibility of performing anonymization on EHR data with reference to their usability in medical decision-making. Furthermore, we summarized the lawful basis for delivering guidance on practical EHR data anonymization. Conclusions This systematic literature mapping study indicates that anonymization of EHR data is theoretically achievable; yet, it requires more research efforts in practical implementations to balance privacy preservation and usability to ensure more reliable health care applications.

show abstract

Section: Resultsmentioning

confidence: 99%

Data Anonymization for Pervasive Health Care: Systematic Literature Mapping Study

Zuo¹,

Watson²,

Budgen³

et al. 2021

JMIR Med Inform

View full text Add to dashboard Cite

show abstract

“…Since the CNN-based model considers both low-and high-frequency components, addressing high-frequency components in the frequency domain plays a significant role [13]. Several recent works have been introduced to solve various problems in the frequency domain instead of in the spatial domain [14,15]. In this paper, we propose a novel data augmentation method by filtering specific patterns in the frequency domain.…”

Section: Frequency Domain Data Augmentationmentioning

confidence: 99%

Frequency Filtering for Data Augmentation in X-Ray Image Classification

Nam

Lee

2021

2021 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

A sufficiently large dataset is needed to solve tasks based on deep learning. However, using medical datasets for deep learning is challenging because their access is limited. In this paper, we propose a novel data augmentation paradigm that can be used in image classification of convolutional neural network-based approaches. As the frequency domain has more unique patterns than the spatial domain, more diverse patterns can be obtained by arbitrarily changing the frequency domain patterns of the given image. First, we assume that the meaningful patterns in the frequency domain are typically distributed in high-intensity regions in the Fourier spectrum. Next, we select the angle to reject meaningful patterns from the Fourier spectrum. Subsequently, we generate a novel mask pattern and remove specific frequency patterns in a fast Fourier transform image. Thereafter, we apply an inverse fast Fourier transform to the rejected frequency image to convert it back to the spatial domain. Our method achieves consistent performance improvements on X-ray images using various backbones. On average, we achieved 3.30%, 7.09%, 7.75%, 8.14%, and 4.91% improvements on the accuracy, precision, recall, F1-score, and AUC over the previous methods.

show abstract

“…Our protocol guarantees that Alice learns only the segmentation results, while Bob learns nothing about the inputs from Alice. The security of the proposed protocol can be easily reduced to that of existing works [3,15], where any attack against BUNET will result in a non-negligible advantage against the ENSEI [3] and Gazelle [15] protocols.…”

Section: Threat Model and Securitymentioning

confidence: 99%

“…HomConv: The HomConv operator obliviously convolve two vectors u and w. A very recent work [3] discovered that, instead of the complex rotate-andaccumulate approach proposed by previous works [15], homomorphic convolution can be performed in the frequency domain, where the only computation needed in the homomorphic domain is the (homomorphic Hadamard product) operator. Therefore, the homomorphic convolution protocol proceeds as follows.…”

Section: The Cryptographic Building Blocksmentioning

confidence: 99%

“…1. First, Alice performs an integer discrete Fourier transform (DFT) (i.e., number theoretic transform in [3]) on u and obtain its frequency-domain representation, û. She simply encrypts this input array into a ciphertext [û] (when ℓ > n, the vector is encrypted into multiple ciphertexts) by running [û] = Enc K (û), where K is the encryption key.…”

Section: The Cryptographic Building Blocksmentioning

confidence: 99%

See 1 more Smart Citation

BUNET: Blind Medical Image Segmentation Based on Secure UNET

Bian

Jiang

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

The strict security requirements placed on medical records by various privacy regulations become major obstacles in the age of big data. To ensure efficient machine learning as a service schemes while protecting data confidentiality, in this work, we propose blind UNET (BUNET), a secure protocol that implements privacy-preserving medical image segmentation based on the UNET architecture. In BUNET, we efficiently utilize cryptographic primitives such as homomorphic encryption and garbled circuits (GC) to design a complete secure protocol for the UNET neural architecture. In addition, we perform extensive architectural search in reducing the computational bottleneck of GCbased secure activation protocols with high-dimensional input data. In the experiment, we thoroughly examine the parameter space of our protocol, and show that we can achieve up to 14x inference time reduction compared to the-state-of-the-art secure inference technique on a baseline architecture with negligible accuracy degradation.

show abstract

ENSEI: Efficient Secure Inference via Frequency-Domain Homomorphic Convolution for Privacy-Preserving Visual Recognition

Cited by 28 publications

References 20 publications

Data Anonymization for Pervasive Health Care: Systematic Literature Mapping Study

Data Anonymization for Pervasive Health Care: Systematic Literature Mapping Study

Frequency Filtering for Data Augmentation in X-Ray Image Classification

BUNET: Blind Medical Image Segmentation Based on Secure UNET

Contact Info

Product

Resources

About