2018
DOI: 10.48550/arxiv.1806.04972
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Unsupervised Detection of Lesions in Brain MRI using constrained adversarial auto-encoders

Abstract: Lesion detection in brain Magnetic Resonance Images (MRI) remains a challenging task. State-of-the-art approaches are mostly based on supervised learning making use of large annotated datasets. Human beings, on the other hand, even nonexperts, can detect most abnormal lesions after seeing a handful of healthy brain images. Replicating this capability of using prior information on the appearance of healthy brain structure to detect lesions can help computers achieve human level abnormality detection, specifical… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
74
0

Year Published

2019
2019
2023
2023

Publication Types

Select...
5
4

Relationship

0
9

Authors

Journals

citations
Cited by 54 publications
(74 citation statements)
references
References 13 publications
0
74
0
Order By: Relevance
“…Quantile regression is a simple yet powerful method for estimating uncertainty both in supervised and unsupervised lesion detection. In the unsupervised framework we used the VAE, a popular model for unsupervised lesion detection (Chen and Konukoglu, 2018;Baur et al, 2018;Pawlowski et al, 2018). VAEs can be used to estimate reconstruction probability instead of reconstruction error for anomaly detection tasks.…”
Section: Discussionmentioning
confidence: 99%
“…Quantile regression is a simple yet powerful method for estimating uncertainty both in supervised and unsupervised lesion detection. In the unsupervised framework we used the VAE, a popular model for unsupervised lesion detection (Chen and Konukoglu, 2018;Baur et al, 2018;Pawlowski et al, 2018). VAEs can be used to estimate reconstruction probability instead of reconstruction error for anomaly detection tasks.…”
Section: Discussionmentioning
confidence: 99%
“…Here, we briefly explain how our DCAE model can serve as a platform for anomaly detection. They trained a reproduction model with enormous data by updating the model weight parameters to minimize the loss function (see e.g., Chen et al 2018;Zimmerer et al 2018;Baur et al 2021). Therefore, the trained model learns the overall structure representation of the training data, but not the uncommon features existing in them.…”
Section: Model Descriptionmentioning
confidence: 99%
“…In comparison, a similar study conducted by [3] consisting of a multitude of algorithms including AnoVAEGAN [4] and f-AnoGANS, obtained a best mean score of 27.8% Dice after post-processing by f-AnoGANS. Before post-processing the best method was Constrained AutoEncoder [8] with a score of 9.7% Dice. An exhaustive list is presented in Table 1.…”
Section: Ms Lesion Segmentation (Ms-seg2015)mentioning
confidence: 99%