English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Unsupervised anomaly segmentation aims to detect patterns that are distinctfrom any patterns processed during training, commonly called abnormal orout-of-distribution patterns, without providing any associated manualsegmentations. Since anomalies during deployment can lead to model failure,detecting the anomaly can enhance the reliability of models, which is valuablein high-risk domains like medical imaging. This paper introduces MaskedModality Cycles with Conditional Diffusion (MMCCD), a method that enablessegmentation of anomalies across diverse patterns in multimodal MRI. The methodis based on two fundamental ideas. First, we propose the use of cyclic modalitytranslation as a mechanism for enabling abnormality detection.Image-translation models learn tissue-specific modality mappings, which arecharacteristic of tissue physiology. Thus, these learned mappings fail totranslate tissues or image patterns that have never been encountered duringtraining, and the error enables their segmentation. Furthermore, we combineimage translation with a masked conditional diffusion model, which attempts to`imagine' what tissue exists under a masked area, further exposing unknownpatterns as the generative model fails to recreate them. We evaluate our methodon a proxy task by training on healthy-looking slices of BraTS2021multi-modality MRIs and testing on slices with tumors. We show that our methodcompares favorably to previous unsupervised approaches based on imagereconstruction and denoising with autoencoders and diffusion models.

Modality Cycles with Masked Conditional Diffusion for Unsupervised Anomaly Segmentation in MRI