Generalizing Unsupervised Anomaly Detection: Towards Unbiased Pathology Screening.
In: (6th International Conference on Medical Imaging with Deep Learning (MIDL), 10-12 July 2023, Vanderbilt Univ, Nashville). 1269 Law St, San Diego, Ca, United States: Jmlr-journal Machine Learning Research, 2023. 39-52 ( ; 227)
möglich sobald bei der ZB eingereicht worden ist.
The main benefit of unsupervised anomaly detection is the ability to identify arbitrary instances of pathologies even in the absence of training labels or sufficient examples of the rare class(es). Even though much work has been done on using auto-encoders (AE) for anomaly detection, there are still two critical challenges to overcome: First, learning compact and detailed representations of the healthy distribution is cumbersome. Second, the majority of unsupervised algorithms are tailored to detect hyperintense lesions on FLAIR brain MR scans. We found that even state-of-the-art (SOTA) AEs fail to detect several classes of non-hyperintense anomalies on T1w brain MRIs, such as brain atrophy, edema, or resections. In this work, we propose reversed AEs (RA) to generate pseudo-healthy reconstructions and localize various brain pathologies. Our method outperformed SOTA methods on T1w brain MRIs, detecting more global anomalies (AUROC increased from 73.1 to 89.4) and local pathologies (detection rate increased from 52.6% to 86.0%).
Altmetric
Weitere Metriken?
Publikationstyp
Artikel: Konferenzbeitrag
Dokumenttyp
Typ der Hochschulschrift
Herausgeber
Korrespondenzautor
Schlagwörter
Unsupervised Anomaly Detection; Pathology Screening
Keywords plus
ISSN (print) / ISBN
2640-3498
e-ISSN
ISBN
Bandtitel
Konferenztitel
6th International Conference on Medical Imaging with Deep Learning (MIDL)
Konferzenzdatum
10-12 July 2023
Konferenzort
Vanderbilt Univ, Nashville
Konferenzband
Quellenangaben
Band: 227,
Heft: ,
Seiten: 39-52
Artikelnummer: ,
Supplement: ,
Reihe
Verlag
Jmlr-journal Machine Learning Research
Verlagsort
1269 Law St, San Diego, Ca, United States
Hochschule
Hochschulort
Fakultät
Veröffentlichungsdatum
0000-00-00
Anmeldedatum
0000-00-00
Anmelder/Inhaber
weitere Inhaber
Anmeldeland
Priorität
Begutachtungsstatus
Förderungen
Helmholtz Association under the joint research school "Munich School for Data Science -MUDS"