Anomaly Detection for Medical Images Using Self-Supervised and Translation-Consistent Features
- 1 July 2021
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Medical Imaging
- Vol. 40 (12), 3641-3651
- https://doi.org/10.1109/tmi.2021.3093883
Abstract
As the labeled anomalous medical images are usually difficult to acquire, especially for rare diseases, the deep learning based methods, which heavily rely on the large amount of labeled data, cannot yield a satisfactory performance. Compared to the anomalous data, the normal images without the need of lesion annotation are much easier to collect. In this paper, we propose an anomaly detection framework, namely $\mathbb {SALAD}$ , extracting $\mathbb {S}$ elf-supervised and tr $\mathbb {A}$ ns $\mathbb {L}$ ation-consistent features for $\mathbb {A}$ nomaly $\mathbb {D}$ etection. The proposed SALAD is a reconstruction-based method, which learns the manifold of normal data through an encode-and-reconstruct translation between image and latent spaces. In particular, two constraints ( i.e. , structure similarity loss and center constraint loss) are proposed to regulate the cross-space ( i.e. , image and feature) translation, which enforce the model to learn translation-consistent and representative features from the normal data. Furthermore, a self-supervised learning module is engaged into our framework to further boost the anomaly detection accuracy by deeply exploiting useful information from the raw normal data. An anomaly score, as a measure to separate the anomalous data from the healthy ones, is constructed based on the learned self-supervised-and-translation-consistent features. Extensive experiments are conducted on optical coherence tomography (OCT) and chest X-ray datasets. The experimental results demonstrate the effectiveness of our approach.Keywords
Funding Information
- Key-Area Research and Development Program of Guangdong Province, China (2018B010111001)
- National Key Research and Development Program of China (2018YFC2000702)
- Scientific and Technical Innovation 2030—New Generation Artificial Intelligence Project (2020AAA0104100)
- China Postdoctoral Science Foundation (2020M680387)
This publication has 37 references indexed in Scilit:
- Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus PhotographsJAMA, 2016
- Unsupervised Learning of Visual Representations by Solving Jigsaw PuzzlesPublished by Springer Science and Business Media LLC ,2016
- Precomputed Real-Time Texture Synthesis with Markovian Generative Adversarial NetworksPublished by Springer Science and Business Media LLC ,2016
- A Discriminative Feature Learning Approach for Deep Face RecognitionPublished by Springer Science and Business Media LLC ,2016
- A Deep Learning Approach for Network Intrusion Detection SystemPublished by European Alliance for Innovation n.o. ,2016
- Unsupervised Visual Representation Learning by Context PredictionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2015
- Anomaly Detection and Localization in Crowded ScenesIEEE Transactions on Pattern Analysis and Machine Intelligence, 2013
- Sparse reconstruction cost for abnormal event detectionPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2011
- Online detection of unusual events in videos via dynamic sparse codingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2011
- Image Quality Assessment: From Error Visibility to Structural SimilarityIEEE Transactions on Image Processing, 2004