Filters








1,653 Hits in 3.2 sec

Unsupervised 3D Brain Anomaly Detection [article]

Jaime Simarro Viana, Ezequiel de la Rosa, Thijs Vande Vyvere, David Robben, Diana M. Sima
2020 arXiv   pre-print
To the best of our knowledge, this study exemplifies the first AD approach that can efficiently handle volumetric data and detect 3D brain anomalies in one single model.  ...  Anomaly detection (AD) is the identification of data samples that do not fit a learned data distribution.  ...  In contrast, unsupervised learning models are capable of discovering patterns from label-free databases. A current challenge in this field is unsupervised anomaly detection (AD).  ... 
arXiv:2010.04717v1 fatcat:mi7nin7i5jgxrjdcflrwx6ecay

Unsupervised Anomaly Detection in 3D Brain MRI using Deep Learning with impured training data [article]

Finn Behrendt, Marcel Bengs, Frederik Rogge, Julia Krüger, Roland Opfer, Alexander Schlaefer
2022 arXiv   pre-print
We study how unhealthy samples within the training data affect anomaly detection performance for brain MRI-scans.  ...  Recently, unsupervised anomaly detection (UAD) methods have shown promising results for this task. These methods rely on training data sets that solely contain healthy samples.  ...  Deep learning for unsupervised anomaly detection follows the concept of identifying abnormalities by learning from a reference data set of healthy data.  ... 
arXiv:2204.05778v1 fatcat:imx3hipiovddfhymf2whqidita

Unsupervised Anomaly Detection in 3D Brain MRI using Deep Learning with Multi-Task Brain Age Prediction [article]

Marcel Bengs, Finn Behrendt, Max-Heinrich Laves, Julia Krüger, Roland Opfer, Alexander Schlaefer
2022
Recently, unsupervised anomaly detection (UAD) in brain MRI with deep learning has shown promising results to provide a quick, initial assessment.  ...  So far, these methods only rely on the visual appearance of healthy brain anatomy for anomaly detection.  ...  We evaluate our methods based on sample level anomaly detection performance, i.e., whether a 3D MRI of a subject is abnormal.  ... 
doi:10.48550/arxiv.2201.13081 fatcat:njh2yy5nwzdpdjwdiwk72m7wwq

MADGAN: unsupervised Medical Anomaly Detection GAN using multiple adjacent brain MRI slice reconstruction [article]

Changhee Han, Leonardo Rundo, Kohei Murao, Tomoyuki Noguchi, Yuki Shimahara, Zoltan Adam Milacski, Saori Koshino, Evis Sala, Hideki Nakayama, Shinichi Satoh
2020 arXiv   pre-print
Therefore, we propose unsupervised Medical Anomaly Detection Generative Adversarial Network (MADGAN), a novel two-step method using GAN-based multiple adjacent brain MRI slice reconstruction to detect  ...  Towards this, unsupervised methods reconstruct a 2D/3D single medical image to detect outliers either in the learned feature space or from high reconstruction loss.  ...  Therefore, this paper proposes unsupervised Medical Anomaly Detection GAN (MADGAN), a novel two-step method using GAN-based multiple adjacent brain MRI slice reconstruction to detect various diseases at  ... 
arXiv:2007.13559v2 fatcat:7g35ohgp6jbifjdlcgozrsfdr4

MADGAN: unsupervised medical anomaly detection GAN using multiple adjacent brain MRI slice reconstruction

Changhee Han, Leonardo Rundo, Kohei Murao, Tomoyuki Noguchi, Yuki Shimahara, Zoltán Ádám Milacski, Saori Koshino, Evis Sala, Hideki Nakayama, Shin'ichi Satoh
2021 BMC Bioinformatics  
We propose unsupervised medical anomaly detection generative adversarial network (MADGAN), a novel two-step method using GAN-based multiple adjacent brain MRI slice reconstruction to detect brain anomalies  ...  Towards this, unsupervised methods reconstruct a 2D/3D single medical image to detect outliers either in the learned feature space or from high reconstruction loss.  ...  However, it is difficult to directly compare our approach with such existing unsupervised anomaly detection methods on 3D medical images since we perform a whole-brain diagnosis (i.e., classification),  ... 
doi:10.1186/s12859-020-03936-1 pmid:33902457 fatcat:jaksh2f2cfcgtkzlzktq3pdhp4

3-Dimensional Deep Learning with Spatial Erasing for Unsupervised Anomaly Segmentation in Brain MRI [article]

Marcel Bengs, Finn Behrendt, Julia Krüger, Roland Opfer, Alexander Schlaefer
2021 arXiv   pre-print
Recently, deep learning methods for unsupervised anomaly detection (UAD) have been proposed for the analysis of brain MRI.  ...  We propose 3D deep learning methods for UAD in brain MRI combined with 3D erasing and demonstrate that 3D methods clearly outperform their 2D counterpart for anomaly segmentation.  ...  Keywords Anomaly · Segmentation · Unsupervised · Brain MRI · 3D Autoencoder 1 Introduction Brain Magnetic Resonance Images (MRIs) allow for three-dimensional (3D) imaging of the brain and are widely used  ... 
arXiv:2109.06540v1 fatcat:5nq3xeh4hrcsrmgsaihwo253cq

Deep Autoencoding Models for Unsupervised Anomaly Segmentation in Brain MR Images [chapter]

Christoph Baur, Benedikt Wiestler, Shadi Albarqouni, Nassir Navab
2019 Lecture Notes in Computer Science  
Previous approaches towards deep unsupervised anomaly detection model patches of normal anatomy with variants of Autoencoders or GANs, and detect anomalies either as outliers in the learned feature space  ...  A plethora of such unsupervised anomaly detection approaches has been made in the medical domain, based on statistical methods, content-based retrieval, clustering and recently also deep learning.  ...  autoencoding models for unsupervised anomaly detection.  ... 
doi:10.1007/978-3-030-11723-8_16 fatcat:7ul5qvn74rhjna5prdmrkemuuq

Three-dimensional deep learning with spatial erasing for unsupervised anomaly segmentation in brain MRI

Marcel Bengs, Finn Behrendt, Julia Krüger, Roland Opfer, Alexander Schlaefer
2021 International Journal of Computer Assisted Radiology and Surgery  
Recently, deep learning methods for unsupervised anomaly detection (UAD) have been proposed for the analysis of brain MRI.  ...  Conclusions We propose 3D deep learning methods for UAD in brain MRI combined with 3D erasing and demonstrate that 3D methods clearly outperform their 2D counterpart for anomaly segmentation.  ...  Discussion We consider the problem of unsupervised anomaly segmentation and propose to learn from entire 3D MRI volumes instead of single 2D MRI.  ... 
doi:10.1007/s11548-021-02451-9 pmid:34251654 fatcat:5u6jnshbsrczljfsacddtv5x54

Leveraging 3d Information In Unsupervised Brain Mri Segmentation

Benjamin Lambert, Maxime Louis, Senan Doyle, Florence Forbes, Michel Dojat, Alan Tucholka
2021 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI)  
To tackle anatomical variability, Unsupervised Anomaly Detection (UAD) methods are proposed, detecting anomalies as outliers of a healthy model learned using a Variational Autoencoder (VAE).  ...  Here, we propose to perform UAD in a 3D fashion and compare 2D and 3D VAEs. As a side contribution, we present a new loss function guarantying a robust training.  ...  To overcome these limitations, Unsupervised Anomaly Detection (UAD) methods are proposed, consisting in the construction of a manifold of healthy scans.  ... 
doi:10.1109/isbi48211.2021.9433894 fatcat:irh7fizdwfh53ahgyelrencuim

Leveraging 3D Information in Unsupervised Brain MRI Segmentation [article]

Benjamin Lambert, Maxime Louis, Senan Doyle, Florence Forbes, Michel Dojat, Alan Tucholka
2021 arXiv   pre-print
To tackle anatomical variability, Unsupervised Anomaly Detection (UAD) methods are proposed, detecting anomalies as outliers of a healthy model learned using a Variational Autoencoder (VAE).  ...  Here, we propose to perform UAD in a 3D fashion and compare 2D and 3D VAEs. As a side contribution, we present a new loss function guarantying a robust training.  ...  To overcome these limitations, Unsupervised Anomaly Detection (UAD) methods are proposed.  ... 
arXiv:2101.10674v1 fatcat:gz64jqhmp5gxvfrsu7nwdb4zou

The MVTec 3D-AD Dataset for Unsupervised 3D Anomaly Detection and Localization [article]

Paul Bergmann, Xin Jin, David Sattlegger, Carsten Steger
2021 arXiv   pre-print
We introduce the first comprehensive 3D dataset for the task of unsupervised anomaly detection and localization.  ...  An initial benchmark of 3D anomaly detection methods on our dataset indicates a considerable room for improvement.  ...  The GAN supervised 3D Brain Anomaly Detection.  ... 
arXiv:2112.09045v1 fatcat:23d67jcu4rcrrbi4762kekqmdi

Anomaly Detection in Medical Imaging – A Mini Review [article]

Maximilian E. Tschuchnig, Michael Gadermayr
2021 arXiv   pre-print
Anomaly detection is one possible methodology that is able to leverage semi-supervised and unsupervised methods to handle medical imaging tasks like classification and segmentation.  ...  further advice on how to approach anomaly detection in medical imaging.  ...  OF CT AND FMRI IMAGE BASED RESULTS OF THE BRAIN OBTAINED BY THE LITERATURE REVIEW Paper Imaging Method Aim Applied Method [21] head CT (3D) anomaly detection (emergency head CTs) 3D cAE [30  ... 
arXiv:2108.11986v1 fatcat:lzffrlo7ovfppiejztkafzlpyq

GAN-based Multiple Adjacent Brain MRI Slice Reconstruction for Unsupervised Alzheimer's Disease Diagnosis [article]

Changhee Han, Leonardo Rundo, Kohei Murao, Zoltán Ádám Milacski, Kazuki Umemoto, Evis Sala, Hideki Nakayama, Shin'ichi Satoh
2020 arXiv   pre-print
Moreover, no study has shown how unsupervised anomaly detection is associated with disease stages.  ...  is fully unsupervised, it should also discover and alert any anomalies including rare disease.  ...  Fig. 3 shows ROC curves and their AUCs of unsupervised anomaly detection.  ... 
arXiv:1906.06114v5 fatcat:23inzriztbe4ngga6kj7kpz4qe

StRegA: Unsupervised Anomaly Detection in Brain MRIs using a Compact Context-encoding Variational Autoencoder [article]

Soumick Chatterjee, Alessandro Sciarra, Max Dünnwald, Pavan Tummala, Shubham Kumar Agrawal, Aishwarya Jauhari, Aman Kalra, Steffen Oeltze-Jafra, Oliver Speck, Andreas Nürnberger
2022 arXiv   pre-print
Hence, a possible solution is an unsupervised anomaly detection (UAD) system that can learn a data distribution from an unlabelled dataset of healthy subjects and then be applied to detect out of distribution  ...  anomalies such as tumours in brain MRIs.  ...  Finally, all the slice-wise 2D masks are stacked together to get the final anomaly mask in 3D. Datasets Training unsupervised anomaly detection models warrant anomaly-free datasets.  ... 
arXiv:2201.13271v2 fatcat:uvvhr4ydnrejvngcwjkrv34zui

Unsupervised Region-Based Anomaly Detection In Brain MRI With Adversarial Image Inpainting

Bao Nguyen, Adam Feldman, Sarath Bethapudi, Andrew Jennings, Chris G. Willcocks
2021 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI)  
In contrast, this paper proposes a fully automatic, unsupervised inpainting-based brain tumour segmentation system for T1-weighted MRI.  ...  First, a deep convolutional neural network (DCNN) is trained to reconstruct missing healthy brain regions.  ...  The evaluation data used [16] was also made open access by the Centre for Clinical Brain Sciences from the University of Edinburgh.  ... 
doi:10.1109/isbi48211.2021.9434115 fatcat:hg2hvrkikbd6vjmikktvletrbm
« Previous Showing results 1 — 15 out of 1,653 results