Unsupervised Anomaly Localization with Structural Feature-Autoencoders

Felix Meissen, Johannes Paetzold, Georgios Kaissis, Daniel Rueckert

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

3 Zitate (Scopus)

Abstract

Unsupervised Anomaly Detection has become a popular method to detect pathologies in medical images as it does not require supervision or labels for training. Most commonly, the anomaly detection model generates a “normal” version of an input image, and the pixel-wise lp -difference of the two is used to localize anomalies. However, large residuals often occur due to imperfect reconstruction of the complex anatomical structures present in most medical images. This method also fails to detect anomalies that are not characterized by large intensity differences to the surrounding tissue. We propose to tackle this problem using a feature-mapping function that transforms the input intensity images into a space with multiple channels where anomalies can be detected along different discriminative feature maps extracted from the original image. We then train an Autoencoder model in this space using structural similarity loss that does not only consider differences in intensity but also in contrast and structure. Our method significantly increases performance on two medical data sets for brain MRI. Code and experiments are available at https://github.com/FeliMe/feature-autoencoder.

OriginalspracheEnglisch
TitelBrainlesion
UntertitelGlioma, Multiple Sclerosis, Stroke and Traumatic Brain Injuries - 8th International Workshop, BrainLes 2022, Held in Conjunction with MICCAI 2022, Revised Selected Papers
Redakteure/-innenSpyridon Bakas, Ujjwal Baid, Bhakti Baheti, Alessandro Crimi, Sylwia Malec, Monika Pytlarz, Maximilian Zenk, Reuben Dorent
Herausgeber (Verlag)Springer Science and Business Media Deutschland GmbH
Seiten14-24
Seitenumfang11
ISBN (Print)9783031338410
DOIs
PublikationsstatusVeröffentlicht - 2023
VeranstaltungProceedings of the 8th International MICCAI Brainlesion Workshop, BrainLes 2022 - Singapore, Singapur
Dauer: 18 Sept. 202222 Sept. 2022

Publikationsreihe

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Band13769 LNCS
ISSN (Print)0302-9743
ISSN (elektronisch)1611-3349

Konferenz

KonferenzProceedings of the 8th International MICCAI Brainlesion Workshop, BrainLes 2022
Land/GebietSingapur
OrtSingapore
Zeitraum18/09/2222/09/22

Fingerprint

Untersuchen Sie die Forschungsthemen von „Unsupervised Anomaly Localization with Structural Feature-Autoencoders“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren