- Keywords: Transformer, Unsupervised Anomaly Segmentation, Anomaly Detection, Neuroimaging, Vector Quantized Variational Autoencoder
- TL;DR: Transformers for Anomaly Detection
- Abstract: Pathological brain appearances may be so heterogeneous as to be intelligible only as anomalies, defined by their deviation from normality rather than any specific pathological characteristic. Amongst the hardest tasks in medical imaging, detecting such anomalies requires models of the normal brain that combine compactness with the expressivity of the complex, long-range interactions that characterise its structural organisation. These are requirements transformers have arguably greater potential to satisfy than other current candidate architectures, but their application has been inhibited by their demands on data and computational resource. Here we combine the latent representation of vector quantised variational autoencoders with an ensemble of autoregressive transformers to enable unsupervised anomaly detection and segmentation defined by deviation from healthy brain imaging data, achievable at low computational cost, within relative modest data regimes. We compare our method to current state-of-the-art approaches across a series of experiments involving synthetic and real pathological lesions. On real lesions, we train our models on 15,000 radiologically normal participants from UK Biobank, and evaluate performance on four different brain MR datasets with small vessel disease, demyelinating lesions, and tumours. We demonstrate superior anomaly detection performance both image-wise and pixel-wise, achievable without post-processing. These results draw attention to the potential of transformers in this most challenging of imaging tasks.
- Registration: I acknowledge that publication of this at MIDL and in the proceedings requires at least one of the authors to register and present the work during the conference.
- Authorship: I confirm that I am the author of this work and that it has not been submitted to another publication before.
- Paper Type: both
- Primary Subject Area: Detection and Diagnosis
- Secondary Subject Area: Unsupervised Learning and Representation Learning