TY - GEN
T1 - Towards practical unsupervised anomaly detection on retinal images
AU - Ouardini, Khalil
AU - Yang, Huijuan
AU - Unnikrishnan, Balagopal
AU - Romain, Manon
AU - Garcin, Camille
AU - Zenati, Houssam
AU - Campbell, J. Peter
AU - Chiang, Michael F.
AU - Kalpathy-Cramer, Jayashree
AU - Chandrasekhar, Vijay
AU - Krishnaswamy, Pavitra
AU - Foo, Chuan Sheng
N1 - Publisher Copyright:
© Springer Nature Switzerland AG 2019.
PY - 2019
Y1 - 2019
N2 - Supervised deep learning approaches provide state-of-the-art performance on medical image classification tasks for disease screening. However, these methods require large labeled datasets that involve resource-intensive expert annotation. Further, disease screening applications have low prevalence of abnormal samples; this class imbalance makes the task more akin to anomaly detection. While the machine learning community has proposed unsupervised deep learning methods for anomaly detection, they have yet to be characterized on medical images where normal vs. anomaly distinctions may be more subtle and variable. In this work, we characterize existing unsupervised anomaly detection methods on retinal fundus images, and find that they require significant fine tuning and offer unsatisfactory performance. We thus propose an efficient and effective transfer-learning based approach for unsupervised anomaly detection. Our method employs a deep convolutional neural network trained on ImageNet as a feature extractor, and subsequently feeds the learned feature representations into an existing unsupervised anomaly detection method. We show that our approach significantly outperforms baselines on two natural image datasets and two retinal fundus image datasets, all with minimal fine-tuning. We further show the ability to leverage very small numbers of labelled anomalies to improve performance. Our work establishes a strong unsupervised baseline for image-based anomaly detection, alongside a flexible and scalable approach for screening applications.
AB - Supervised deep learning approaches provide state-of-the-art performance on medical image classification tasks for disease screening. However, these methods require large labeled datasets that involve resource-intensive expert annotation. Further, disease screening applications have low prevalence of abnormal samples; this class imbalance makes the task more akin to anomaly detection. While the machine learning community has proposed unsupervised deep learning methods for anomaly detection, they have yet to be characterized on medical images where normal vs. anomaly distinctions may be more subtle and variable. In this work, we characterize existing unsupervised anomaly detection methods on retinal fundus images, and find that they require significant fine tuning and offer unsatisfactory performance. We thus propose an efficient and effective transfer-learning based approach for unsupervised anomaly detection. Our method employs a deep convolutional neural network trained on ImageNet as a feature extractor, and subsequently feeds the learned feature representations into an existing unsupervised anomaly detection method. We show that our approach significantly outperforms baselines on two natural image datasets and two retinal fundus image datasets, all with minimal fine-tuning. We further show the ability to leverage very small numbers of labelled anomalies to improve performance. Our work establishes a strong unsupervised baseline for image-based anomaly detection, alongside a flexible and scalable approach for screening applications.
KW - Anomaly detection
KW - Retinal images
KW - Transfer learning
KW - Unsupervised deep learning
UR - http://www.scopus.com/inward/record.url?scp=85075642341&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85075642341&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-33391-1_26
DO - 10.1007/978-3-030-33391-1_26
M3 - Conference contribution
AN - SCOPUS:85075642341
SN - 9783030333904
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 225
EP - 234
BT - Domain Adaptation and Representation Transfer and Medical Image Learning with Less Labels and Imperfect Data First MICCAI Workshop, DART 2019 and First International Workshop, MIL3ID 2019 Shenzhen, Held in Conjunction with MICCAI 2019 Shenzhen, 2019 Proceedings
A2 - Wang, Qian
A2 - Milletari, Fausto
A2 - Rieke, Nicola
A2 - Nguyen, Hien V.
A2 - Roysam, Badri
A2 - Albarqouni, Shadi
A2 - Cardoso, M. Jorge
A2 - Xu, Ziyue
A2 - Kamnitsas, Konstantinos
A2 - Patel, Vishal
A2 - Jiang, Steve
A2 - Zhou, Kevin
A2 - Luu, Khoa
A2 - Le, Ngan
PB - Springer
T2 - 1st MICCAI Workshop on Domain Adaptation and Representation Transfer, DART 2019, and the 1st International Workshop on Medical Image Learning with Less Labels and Imperfect Data, MIL3ID 2019, held in conjunction with 22nd International Conference on Medical Image Computing and Computer- Assisted Intervention, MICCAI 2019
Y2 - 13 October 2019 through 17 October 2019
ER -