Data privacy protection domain adaptation by roughing and finishing stage

The automatic segmentation of organs or tissues is crucial for early diagnosis and treatment. Existing deep learning methods either need massive annotation data or use Unsupervised Domain Adaptation (UDA) approaches with labeled source domain data to train a model for unlabeled target domain data. T...

Full description

Saved in:
Bibliographic Details
Main Authors: Yuan, Liqiang, Erdt, Marius, Li, Ruilin, Siyal, Mohammed Yakoob
Other Authors: School of Electrical and Electronic Engineering
Format: Article
Language:English
Published: 2023
Subjects:
Online Access:https://hdl.handle.net/10356/172253
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-172253
record_format dspace
spelling sg-ntu-dr.10356-1722532023-12-04T04:51:18Z Data privacy protection domain adaptation by roughing and finishing stage Yuan, Liqiang Erdt, Marius Li, Ruilin Siyal, Mohammed Yakoob School of Electrical and Electronic Engineering Fraunhofer Singapore Engineering::Electrical and electronic engineering Unsupervised Domain Adaptation Model Generalization The automatic segmentation of organs or tissues is crucial for early diagnosis and treatment. Existing deep learning methods either need massive annotation data or use Unsupervised Domain Adaptation (UDA) approaches with labeled source domain data to train a model for unlabeled target domain data. The methods mentioned above require accessing the data of the source domain. However, in medical imaging, source domain data from a hospital is usually restricted due to patient data privacy regulations. Therefore, it is crucial to perform privacy-preserving domain adaptation, which simultaneously improves the model’s performance on target domain data and preserves the privacy of source domain data. Aiming to achieve Source Free Unsupervised Domain Adaptation (SFUDA), we propose a two-stage framework Roughing and Finishing to extract the relevant features and knowledge from the pre-trained model and protect the patient’s privacy. Specifically, in the Roughing stage, batch statistics are updated using unlabelled target domain data to alleviate feature distribution shifts. Then, a non-parametric weighted entropy-minimization is used to reduce the domain gap further. In the Finishing stage, we aim to enhance the model’s generalization ability by increasing the noise robustness of the feature representation, leading to a more generalizable target feature representation. Besides, we propose pseudo-label training based on a confidence-weighted feature distance to utilize the knowledge from confident samples. The proposed method is evaluated on two medical domain transfer challenges, (1) MRI to CT liver segmentation and (2) cross-domain fundus image segmentation. Extensive experiments and ablation studies demonstrate the superiority of the proposed methods over state-of-the-art SFUDA methods by a large margin. The proposed method outperforms previous UDA approaches that access the source domain data. National Research Foundation (NRF) This research is supported by the National Research Foundation, Singapore under its International Research Centres in Singapore Funding Initiative. 2023-12-04T04:51:18Z 2023-12-04T04:51:18Z 2023 Journal Article Yuan, L., Erdt, M., Li, R. & Siyal, M. Y. (2023). Data privacy protection domain adaptation by roughing and finishing stage. Visual Computer. https://dx.doi.org/10.1007/s00371-023-02794-1 0178-2789 https://hdl.handle.net/10356/172253 10.1007/s00371-023-02794-1 2-s2.0-85148867769 en Visual Computer © 2023 The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature. All rights reserved.
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Engineering::Electrical and electronic engineering
Unsupervised Domain Adaptation
Model Generalization
spellingShingle Engineering::Electrical and electronic engineering
Unsupervised Domain Adaptation
Model Generalization
Yuan, Liqiang
Erdt, Marius
Li, Ruilin
Siyal, Mohammed Yakoob
Data privacy protection domain adaptation by roughing and finishing stage
description The automatic segmentation of organs or tissues is crucial for early diagnosis and treatment. Existing deep learning methods either need massive annotation data or use Unsupervised Domain Adaptation (UDA) approaches with labeled source domain data to train a model for unlabeled target domain data. The methods mentioned above require accessing the data of the source domain. However, in medical imaging, source domain data from a hospital is usually restricted due to patient data privacy regulations. Therefore, it is crucial to perform privacy-preserving domain adaptation, which simultaneously improves the model’s performance on target domain data and preserves the privacy of source domain data. Aiming to achieve Source Free Unsupervised Domain Adaptation (SFUDA), we propose a two-stage framework Roughing and Finishing to extract the relevant features and knowledge from the pre-trained model and protect the patient’s privacy. Specifically, in the Roughing stage, batch statistics are updated using unlabelled target domain data to alleviate feature distribution shifts. Then, a non-parametric weighted entropy-minimization is used to reduce the domain gap further. In the Finishing stage, we aim to enhance the model’s generalization ability by increasing the noise robustness of the feature representation, leading to a more generalizable target feature representation. Besides, we propose pseudo-label training based on a confidence-weighted feature distance to utilize the knowledge from confident samples. The proposed method is evaluated on two medical domain transfer challenges, (1) MRI to CT liver segmentation and (2) cross-domain fundus image segmentation. Extensive experiments and ablation studies demonstrate the superiority of the proposed methods over state-of-the-art SFUDA methods by a large margin. The proposed method outperforms previous UDA approaches that access the source domain data.
author2 School of Electrical and Electronic Engineering
author_facet School of Electrical and Electronic Engineering
Yuan, Liqiang
Erdt, Marius
Li, Ruilin
Siyal, Mohammed Yakoob
format Article
author Yuan, Liqiang
Erdt, Marius
Li, Ruilin
Siyal, Mohammed Yakoob
author_sort Yuan, Liqiang
title Data privacy protection domain adaptation by roughing and finishing stage
title_short Data privacy protection domain adaptation by roughing and finishing stage
title_full Data privacy protection domain adaptation by roughing and finishing stage
title_fullStr Data privacy protection domain adaptation by roughing and finishing stage
title_full_unstemmed Data privacy protection domain adaptation by roughing and finishing stage
title_sort data privacy protection domain adaptation by roughing and finishing stage
publishDate 2023
url https://hdl.handle.net/10356/172253
_version_ 1784855587063857152