Self-supervised pseudo multi-class pre-training for unsupervised anomaly detection and segmentation in medical images

Unsupervised anomaly detection (UAD) methods are trained with normal (or healthy) images only, but during testing, they are able to classify normal and abnormal (or disease) images. UAD is an important medical image analysis (MIA) method to be applied in disease screening problems because the traini...

Full description

Saved in:
Bibliographic Details
Main Authors: TIAN, Yu, LIU, Fengbei, PANG, Guansong, CHEN, Yuanhong, LIU, Yuyuan, VERJANS, Johan W., SINGH, Rajvinder, CARNEIRO, Gustavo
Format: text
Language:English
Published: Institutional Knowledge at Singapore Management University 2023
Subjects:
Online Access:https://ink.library.smu.edu.sg/sis_research/8142
https://ink.library.smu.edu.sg/context/sis_research/article/9145/viewcontent/Self_supervised_Pseudo_Medical_sv.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Singapore Management University
Language: English
id sg-smu-ink.sis_research-9145
record_format dspace
spelling sg-smu-ink.sis_research-91452023-09-14T08:19:54Z Self-supervised pseudo multi-class pre-training for unsupervised anomaly detection and segmentation in medical images TIAN, Yu LIU, Fengbei PANG, Guansong CHEN, Yuanhong LIU, Yuyuan VERJANS, Johan W. SINGH, Rajvinder CARNEIRO, Gustavo Unsupervised anomaly detection (UAD) methods are trained with normal (or healthy) images only, but during testing, they are able to classify normal and abnormal (or disease) images. UAD is an important medical image analysis (MIA) method to be applied in disease screening problems because the training sets available for those problems usually contain only normal images. However, the exclusive reliance on normal images may result in the learning of ineffective low-dimensional image representations that are not sensitive enough to detect and segment unseen abnormal lesions of varying size, appearance, and shape. Pre-training UAD methods with self-supervised learning, based on computer vision techniques, can mitigate this challenge, but they are sub-optimal because they do not explore domain knowledge for designing the pretext tasks, and their contrastive learning losses do not try to cluster the normal training images, which may result in a sparse distribution of normal images that is ineffective for anomaly detection. In this paper, we propose a new self-supervised pre-training method for MIA UAD applications, named Pseudo Multi-class Strong Augmentation via Contrastive Learning (PMSACL). PMSACL consists of a novel optimisation method that contrasts a normal image class from multiple pseudo classes of synthesised abnormal images, with each class enforced to form a dense cluster in the feature space. In the experiments, we show that our PMSACL pre-training improves the accuracy of SOTA UAD methods on many MIA benchmarks using colonoscopy, fundus screening and Covid-19 Chest X-ray datasets. 2023-12-01T08:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/8142 info:doi/10.1016/j.media.2023.102930 https://ink.library.smu.edu.sg/context/sis_research/article/9145/viewcontent/Self_supervised_Pseudo_Medical_sv.pdf http://creativecommons.org/licenses/by-nc-nd/4.0/ Research Collection School Of Computing and Information Systems eng Institutional Knowledge at Singapore Management University Anomaly segmentation Colonoscopy Covid-19 Fundus image Lesion segmentation One-class classification Self-supervised learning Unsupervised anomaly detection Databases and Information Systems Graphics and Human Computer Interfaces Medical Sciences
institution Singapore Management University
building SMU Libraries
continent Asia
country Singapore
Singapore
content_provider SMU Libraries
collection InK@SMU
language English
topic Anomaly segmentation
Colonoscopy
Covid-19
Fundus image
Lesion segmentation
One-class classification
Self-supervised learning
Unsupervised anomaly detection
Databases and Information Systems
Graphics and Human Computer Interfaces
Medical Sciences
spellingShingle Anomaly segmentation
Colonoscopy
Covid-19
Fundus image
Lesion segmentation
One-class classification
Self-supervised learning
Unsupervised anomaly detection
Databases and Information Systems
Graphics and Human Computer Interfaces
Medical Sciences
TIAN, Yu
LIU, Fengbei
PANG, Guansong
CHEN, Yuanhong
LIU, Yuyuan
VERJANS, Johan W.
SINGH, Rajvinder
CARNEIRO, Gustavo
Self-supervised pseudo multi-class pre-training for unsupervised anomaly detection and segmentation in medical images
description Unsupervised anomaly detection (UAD) methods are trained with normal (or healthy) images only, but during testing, they are able to classify normal and abnormal (or disease) images. UAD is an important medical image analysis (MIA) method to be applied in disease screening problems because the training sets available for those problems usually contain only normal images. However, the exclusive reliance on normal images may result in the learning of ineffective low-dimensional image representations that are not sensitive enough to detect and segment unseen abnormal lesions of varying size, appearance, and shape. Pre-training UAD methods with self-supervised learning, based on computer vision techniques, can mitigate this challenge, but they are sub-optimal because they do not explore domain knowledge for designing the pretext tasks, and their contrastive learning losses do not try to cluster the normal training images, which may result in a sparse distribution of normal images that is ineffective for anomaly detection. In this paper, we propose a new self-supervised pre-training method for MIA UAD applications, named Pseudo Multi-class Strong Augmentation via Contrastive Learning (PMSACL). PMSACL consists of a novel optimisation method that contrasts a normal image class from multiple pseudo classes of synthesised abnormal images, with each class enforced to form a dense cluster in the feature space. In the experiments, we show that our PMSACL pre-training improves the accuracy of SOTA UAD methods on many MIA benchmarks using colonoscopy, fundus screening and Covid-19 Chest X-ray datasets.
format text
author TIAN, Yu
LIU, Fengbei
PANG, Guansong
CHEN, Yuanhong
LIU, Yuyuan
VERJANS, Johan W.
SINGH, Rajvinder
CARNEIRO, Gustavo
author_facet TIAN, Yu
LIU, Fengbei
PANG, Guansong
CHEN, Yuanhong
LIU, Yuyuan
VERJANS, Johan W.
SINGH, Rajvinder
CARNEIRO, Gustavo
author_sort TIAN, Yu
title Self-supervised pseudo multi-class pre-training for unsupervised anomaly detection and segmentation in medical images
title_short Self-supervised pseudo multi-class pre-training for unsupervised anomaly detection and segmentation in medical images
title_full Self-supervised pseudo multi-class pre-training for unsupervised anomaly detection and segmentation in medical images
title_fullStr Self-supervised pseudo multi-class pre-training for unsupervised anomaly detection and segmentation in medical images
title_full_unstemmed Self-supervised pseudo multi-class pre-training for unsupervised anomaly detection and segmentation in medical images
title_sort self-supervised pseudo multi-class pre-training for unsupervised anomaly detection and segmentation in medical images
publisher Institutional Knowledge at Singapore Management University
publishDate 2023
url https://ink.library.smu.edu.sg/sis_research/8142
https://ink.library.smu.edu.sg/context/sis_research/article/9145/viewcontent/Self_supervised_Pseudo_Medical_sv.pdf
_version_ 1779157180252946432