Presentation 2023-11-14
Medical image diagnosis support system with image anonymization based on deep learning techniques
Katsuto Iwai, Ryuunosuke Kounosu, Hirokazu Nosato, Yuu Nakajima,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) When medical imaging AI models are hosted on cloud service there is a risk of sensitive medical images being leaked when transmitted over the Internet. In order to overcome this problem, this study proposes an image diagnosis support system with an image anonymization mechanism that irreversibly transforms images and achieves lesion classification. In the proposed approach, medical images are not directly fed to the classification AI model. Instead, a Variational Auto-Encoder (VAE) mechanism is used to extract anonymized features that can be classified by a later classification AI model. Specifically, during the training of the VAE and the classification model, the classification error is backpropagated to teach the VAE to only reconstruct necessary information for classification as features. This study demonstrates the effectiveness of the proposed method for classification and anonymization by utilizing the MNIST dataset, the Fashion-MNIST dataset, the CIFAR-10 dataset and clinical cystoscope images.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Image Anonymization / machine learning / medical images / cloud service / diagnosis support
Paper # MICT2023-30,MI2023-23
Date of Issue 2023-11-07 (MICT, MI)

Conference Information
Committee MI / MICT
Conference Date 2023/11/14(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Ryo Haraguchi(Univ. of Hyogo) / Hirokazu Tanaka(Hiroshima City Univ.)
Vice Chair Hideaki Haneishi(Chiba Univ.) / Takayuki Kitasaka(Aichi Inst. of Tech.) / Chika Sugimoto(Yokohama National Univ.) / Daisuke Anzai(Nagoya Inst. of Tech.)
Secretary Hideaki Haneishi(Yamaguchi Univ.) / Takayuki Kitasaka(NAIST) / Chika Sugimoto(Okayama Pref. Univ.) / Daisuke Anzai(Hiroshima City Univ)
Assistant Takeshi Hara(Gifu Univ.) / Kenichi Morooka(Okayama Univ.) / Dairoku Muramatsu(Univ. of Electro & Comm.) / Natsuki Nakayama(Nagoya Univ.) / Ami Tanaka(Ritsumeikan Univ.) / Kun Li(Kagawa Univ.)

Paper Information
Registration To Technical Committee on Medical Imaging / Technical Committee on Healthcare and Medical Information Communication Technology
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Medical image diagnosis support system with image anonymization based on deep learning techniques
Sub Title (in English)
Keyword(1) Image Anonymization
Keyword(2) machine learning
Keyword(3) medical images
Keyword(4) cloud service
Keyword(5) diagnosis support
1st Author's Name Katsuto Iwai
1st Author's Affiliation Toho University/National Institute of Advanced Industrial Science and Technology(Toho Univ./AIST)
2nd Author's Name Ryuunosuke Kounosu
2nd Author's Affiliation Toho University/National Institute of Advanced Industrial Science and Technology(Toho Univ./AIST)
3rd Author's Name Hirokazu Nosato
3rd Author's Affiliation National Institute of Advanced Industrial Science and Technology(AIST)
4th Author's Name Yuu Nakajima
4th Author's Affiliation Toho University(Toho Univ.)
Date 2023-11-14
Paper # MICT2023-30,MI2023-23
Volume (vol) vol.123
Number (no) MICT-256,MI-257
Page pp.pp.21-24(MICT), pp.21-24(MI),
#Pages 4
Date of Issue 2023-11-07 (MICT, MI)