Constrained contrastive distribution learning for unsupervised anomaly detection and localisation in medical images
Unsupervised anomaly detection (UAD) learns one-class classifiers exclusively with normal (i.e., healthy) images to detect any abnormal (i.e., unhealthy) samples that do not conform to the expected normal patterns. UAD has two main advantages over its fully supervised counterpart. Firstly, it is abl...
Saved in:
Main Authors: | , , , , , , |
---|---|
Format: | text |
Language: | English |
Published: |
Institutional Knowledge at Singapore Management University
2021
|
Subjects: | |
Online Access: | https://ink.library.smu.edu.sg/sis_research/7035 https://ink.library.smu.edu.sg/context/sis_research/article/8038/viewcontent/521400_1_En_Print.indd.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-8038 |
---|---|
record_format |
dspace |
spelling |
sg-smu-ink.sis_research-80382023-08-10T00:29:25Z Constrained contrastive distribution learning for unsupervised anomaly detection and localisation in medical images TIAN, Yu PANG, Guansong LIU, Fengbei CHEN, Yuanhong SHIN, Seon Ho VERJANS, Johan W. SINGH, Rajvinder Unsupervised anomaly detection (UAD) learns one-class classifiers exclusively with normal (i.e., healthy) images to detect any abnormal (i.e., unhealthy) samples that do not conform to the expected normal patterns. UAD has two main advantages over its fully supervised counterpart. Firstly, it is able to directly leverage large datasets available from health screening programs that contain mostly normal image samples, avoiding the costly manual labelling of abnormal samples and the subsequent issues involved in training with extremely class-imbalanced data. Further, UAD approaches can potentially detect and localise any type of lesions that deviate from the normal patterns. One significant challenge faced by UAD methods is how to learn effective low-dimensional image representations to detect and localise subtle abnormalities, generally consisting of small lesions. To address this challenge, we propose a novel self-supervised representation learning method, called Constrained Contrastive Distribution learning for anomaly detection (CCD), which learns fine-grained feature representations by simultaneously predicting the distribution of augmented data and image contexts using contrastive learning with pretext constraints. The learned representations can be leveraged to train more anomaly-sensitive detection models. Extensive experiment results show that our method outperforms current state-of-the-art UAD approaches on three different colonoscopy and fundus screening datasets. Our code is available at https://github.com/tianyu0207/CCD. 2021-10-01T07:00:00Z text application/pdf https://ink.library.smu.edu.sg/sis_research/7035 info:doi/10.1007/978-3-030-87240-3_13 https://ink.library.smu.edu.sg/context/sis_research/article/8038/viewcontent/521400_1_En_Print.indd.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 detection Unsupervised learning Lesion detection and segmentation Self-supervised pre-training Colonoscopy Artificial Intelligence and Robotics Graphics and Human Computer Interfaces Health Information Technology |
institution |
Singapore Management University |
building |
SMU Libraries |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
SMU Libraries |
collection |
InK@SMU |
language |
English |
topic |
Anomaly detection Unsupervised learning Lesion detection and segmentation Self-supervised pre-training Colonoscopy Artificial Intelligence and Robotics Graphics and Human Computer Interfaces Health Information Technology |
spellingShingle |
Anomaly detection Unsupervised learning Lesion detection and segmentation Self-supervised pre-training Colonoscopy Artificial Intelligence and Robotics Graphics and Human Computer Interfaces Health Information Technology TIAN, Yu PANG, Guansong LIU, Fengbei CHEN, Yuanhong SHIN, Seon Ho VERJANS, Johan W. SINGH, Rajvinder Constrained contrastive distribution learning for unsupervised anomaly detection and localisation in medical images |
description |
Unsupervised anomaly detection (UAD) learns one-class classifiers exclusively with normal (i.e., healthy) images to detect any abnormal (i.e., unhealthy) samples that do not conform to the expected normal patterns. UAD has two main advantages over its fully supervised counterpart. Firstly, it is able to directly leverage large datasets available from health screening programs that contain mostly normal image samples, avoiding the costly manual labelling of abnormal samples and the subsequent issues involved in training with extremely class-imbalanced data. Further, UAD approaches can potentially detect and localise any type of lesions that deviate from the normal patterns. One significant challenge faced by UAD methods is how to learn effective low-dimensional image representations to detect and localise subtle abnormalities, generally consisting of small lesions. To address this challenge, we propose a novel self-supervised representation learning method, called Constrained Contrastive Distribution learning for anomaly detection (CCD), which learns fine-grained feature representations by simultaneously predicting the distribution of augmented data and image contexts using contrastive learning with pretext constraints. The learned representations can be leveraged to train more anomaly-sensitive detection models. Extensive experiment results show that our method outperforms current state-of-the-art UAD approaches on three different colonoscopy and fundus screening datasets. Our code is available at https://github.com/tianyu0207/CCD. |
format |
text |
author |
TIAN, Yu PANG, Guansong LIU, Fengbei CHEN, Yuanhong SHIN, Seon Ho VERJANS, Johan W. SINGH, Rajvinder |
author_facet |
TIAN, Yu PANG, Guansong LIU, Fengbei CHEN, Yuanhong SHIN, Seon Ho VERJANS, Johan W. SINGH, Rajvinder |
author_sort |
TIAN, Yu |
title |
Constrained contrastive distribution learning for unsupervised anomaly detection and localisation in medical images |
title_short |
Constrained contrastive distribution learning for unsupervised anomaly detection and localisation in medical images |
title_full |
Constrained contrastive distribution learning for unsupervised anomaly detection and localisation in medical images |
title_fullStr |
Constrained contrastive distribution learning for unsupervised anomaly detection and localisation in medical images |
title_full_unstemmed |
Constrained contrastive distribution learning for unsupervised anomaly detection and localisation in medical images |
title_sort |
constrained contrastive distribution learning for unsupervised anomaly detection and localisation in medical images |
publisher |
Institutional Knowledge at Singapore Management University |
publishDate |
2021 |
url |
https://ink.library.smu.edu.sg/sis_research/7035 https://ink.library.smu.edu.sg/context/sis_research/article/8038/viewcontent/521400_1_En_Print.indd.pdf |
_version_ |
1779156842908221440 |