VECTOR-QUANTIZED LATENT FLOWS FOR MEDICAL IMAGE SYNTHESIS AND OUT-OF-DISTRIBUTION DETECTION

被引：0

作者：

Khader, Firas ^{[1
]}

Mueller-Franzes, Gustav ^{[1
]}

Arasteh, Soroosh Tayebi ^{[1
]}

Han, Tianyu ^{[2
]}

Kather, Jakob Nikolas ^{[3
,4
,5
,6
]}

Stegmaier, Johannes ^{[7
]}

Nebelung, Sven ^{[1
]}

Truhn, Daniel ^{[1
]}

机构：

[1] Univ Hosp Aachen, Dept Diagnost & Intervent Radiol, Aachen, Germany

[2] Rhein Westfal TH Aachen, Phys Mol Imaging Syst, Expt Mol Imaging, Aachen, Germany

[3] Univ Leeds, Leeds Inst Med Res St Jamess, Pathol & Data Analyt, Leeds, W Yorkshire, England

[4] German Canc Res Ctr, German Canc Consortium DKTK, Heidelberg, Germany

[5] Univ Heidelberg Hosp, Natl Ctr Tumor Dis NCT, Med Oncol, Heidelberg, Germany

[6] Tech Univ Dresden, Med Fac Carl Gustav Carus, Else Kroener Fresenius Ctr Digital Hlth, Dresden, Germany

[7] Rhein Westfal TH Aachen, Inst Imaging & Comp Vis, Aachen, Germany

来源：

2023 IEEE 20TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING, ISBI | 2023年

关键词：

Out-of-Distribution Detection; Generative Model; Medical Support;

D O I：

10.1109/ISBI53787.2023.10230460

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

We present an innovative method that allows for simultaneous out-of-distribution detection and image generation by encoding images in the latent space of a vector-quantized autoencoder and using normalizing flow models. The technique is demonstrated on a medical dataset of knee radiographs and can be used to relieve clinical radiologists of tedious tasks of quality control while simultaneously guiding radiologic technologists to improved and standardized image quality during image acquisition.

引用

页数：5

共 50 条

[11] Adequate structuring of the latent space for easy classification and out-of-distribution detection
Ossonce, Maxime
Duhamel, Pierre
Alberge, Florence
32ND EUROPEAN SIGNAL PROCESSING CONFERENCE, EUSIPCO 2024, 2024, : 1776 - 1780
[12] MIM-OOD: Generative Masked Image Modelling for Out-of-Distribution Detection in Medical Images
Marimont, Sergio Naval
Siomos, Vasilis
Tarroni, Giacomo
DEEP GENERATIVE MODELS, DGM4MICCAI 2023, 2024, 14533 : 35 - 44
[13] Information Bottleneck-Based Feature Weighting for Enhanced Medical Image Out-of-Distribution Detection
Schott, Brayden
Klanecek, Zan
Deatsch, Alison
Santoro-Fernandes, Victor
Francken, Thomas
Perlman, Scott
Jeraj, Robert
UNCERTAINTY FOR SAFE UTILIZATION OF MACHINE LEARNING IN MEDICAL IMAGING, UNSURE 2024, 2025, 15167 : 128 - 137
[14] Benchmarking Image Classifiers for Physical Out-of-Distribution Examples Detection
Ojaswee
Agarwal, Akshay
Ratha, Nalini
2023 IEEE/CVF INTERNATIONAL CONFERENCE ON COMPUTER VISION WORKSHOPS, ICCVW, 2023, : 4429 - 4437
[15] An Efficient Data Augmentation Network for Out-of-Distribution Image Detection
Lin, Cheng-Hung
Lin, Cheng-Shian
Chou, Po-Yung
Hsu, Chen-Chien
IEEE ACCESS, 2021, 9 : 35313 - 35323
[16] On the Learnability of Out-of-distribution Detection
Fang, Zhen
Li, Yixuan
Liu, Feng
Han, Bo
Lu, Jie
Journal of Machine Learning Research, 2024, 25
[17] Normalizing Flows for Out-of-Distribution Detection: Application to Coronary Artery Segmentation
Ciusdel, Costin Florian
Itu, Lucian Mihai
Cimen, Serkan
Wels, Michael
Schwemmer, Chris
Fortner, Philipp
Seitz, Sebastian
Andre, Florian
Buss, Sebastian Johannes
Sharma, Puneet
Rapaka, Saikiran
APPLIED SCIENCES-BASEL, 2022, 12 (08):
[18] Codebook Transfer with Part-of-Speech for Vector-Quantized Image Modeling
Zhang, Baoquan
Wang, Huaibin
Luo, Chuyao
Li, Xutao
Liang, Guotao
Ye, Yunming
Qi, Xiaochen
He, Yao
2024 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR 2024, 2024, : 7757 - 7766
[19] Attention Masking for Improved Near Out-of-Distribution Image Detection
Sim, Minho
Lee, Jongwhoa
Choi, Ho-Jin
2023 IEEE INTERNATIONAL CONFERENCE ON BIG DATA AND SMART COMPUTING, BIGCOMP, 2023, : 195 - 202
[20] Leveraging Vector-Quantized Variational Autoencoder Inner Metrics for Anomaly Detection
Gangloff, Hugo
Pham, Minh-Tan
Courtrai, Luc
Lefevre, Sebastien
2022 26TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR), 2022, : 435 - 441

← 1 2 3 4 5 →