Unsupervised domain adaptation for the automated segmentation of neuroanatomy in MRI: a deep learning approach

Novosad, Philip; Fonov, Vladimir; Collins, D. Louis

doi:10.1101/845537

Cited by 4 publications

(3 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To this end, fine-tuning the freelyavailable learned model onto any new MR unit will likely be necessary (32). Unsupervised domain adaptation techniques may be an alternative approach allowing to implement our model onto a new image domain without the need for paired sets (33). Second, patients imaged in the intermediate known onset time window (i.e., 4.5-12h) were underrepresented (namely, 10% of all patients imaged within 12h), as in previous studies (29,34).…”

Section: Discussionmentioning

confidence: 99%

Synthetic FLAIR as a Substitute for FLAIR Sequence in Acute Ischemic Stroke

Benzakoun¹,

Deslys²,

Legrand³

et al. 2022

Radiology

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Synthetic FLAIR as a Substitute for FLAIR Sequence in Acute Ischemic Stroke

Benzakoun¹,

Deslys²,

Legrand³

et al. 2022

Radiology

View full text Add to dashboard Cite

show abstract

“…DASiS solutions have been designed for MRI segmentation of liver and kidney (Valindria et al, 2018), neuroanatomy (Novosad et al, 2019), retinal vessel (Huang et al, 2020b), white matter hyper-intensities (Orbes-Arteaga et al, 2019), and multiple sclerosis lesions (Ackaouy et al, 2020). Furthermore, Bermúdez-Chacón et al (2018) apply DASiS to microscopic image segmentation; Dou et al (2018) and Jiang et al (2018) perform adaptation between CT and MRI images for cardiac structure segmentation and for lung cancer segmentation, respectively.…”

Section: Medical Image Segmentationmentioning

confidence: 99%

Semantic Image Segmentation: Two Decades of Research

Volpi¹,

Chidlovskii²

2022

View full text Add to dashboard Cite

Semantic image segmentation (SiS) plays a fundamental role in a broad variety of computer vision applications, providing key information for the global understanding of an image. This survey is an effort to summarize two decades of research in the field of SiS, where we propose a literature review of solutions starting from early historical methods followed by an overview of more recent deep learning methods including the latest trend of using transformers. We complement the review by discussing particular cases of the weak supervision and side machine learning techniques that can be used to improve the semantic segmentation such as curriculum, incremental or self-supervised learning.State-of-the-art SiS models rely on a large amount of annotated samples, which are more expensive to obtain than labels for tasks such as image classification. Since unlabeled data is instead significantly cheaper to obtain, it is not surprising that Unsupervised Domain Adaptation (UDA) reached a broad success within the semantic segmentation community. Therefore, a second core contribution of this book is to summarize five years of a rapidly growing field, Domain Adaptation for Semantic Image Segmentation (DASiS) which embraces the importance of semantic segmentation itself and a critical need of adapting segmentation models to new environments. In addition to providing a comprehensive survey on DASiS techniques, we unveil also newer trends

show abstract

“…UDA can be used for multi-site segmentation of MS lesions from MRI (Ackaouy et al 2020). Furthermore, extensive data augmentation of MR images has been shown to improve segmentation results (Novosad, Fonov, and Collins 2019;Billot et al 2020). Another UDA approach, based on domainadversarial neural networks (Ganin et al 2016), is used for multi-site brain lesion segmentation (Kamnitsas et al 2017).…”

Section: Related Workmentioning

confidence: 99%

Learn to Ignore: Domain Adaptation for Multi-Site MRI Analysis

Wolleb¹,

Sandkühler²,

Bieder³

et al. 2021

Preprint

View full text Add to dashboard Cite

Limited availability of large image datasets is a major issue in the development of accurate and generalizable machine learning methods in medicine. The limitations in the amount of data are mainly due to the use of different acquisition protocols, different hardware, and data privacy. At the same time, training a classification model on a small dataset leads to a poor generalization quality of the model. To overcome this issue, a combination of various image datasets of different provenance is often used, e.g., multi-site studies. However, if an additional dataset does not include all classes of the task, the learning of the classification model can be biased to the device or place of acquisition. This is especially the case for Magnetic Resonance (MR) images, where different MR scanners introduce a bias that limits the performance of the model. In this paper, we present a novel method that learns to ignore the scanner-related features present in the images, while learning features relevant for the classification task. We focus on a real-world scenario, where only a small dataset provides images of all classes. We exploit this circumstance by introducing specific additional constraints on the latent space, which lead the focus on disease-related rather than scannerspecific features. Our method Learn to Ignore outperforms state-of-the-art domain adaptation methods on a multi-site MRI dataset on a classification task between Multiple Sclerosis patients and healthy subjects.

show abstract

Unsupervised domain adaptation for the automated segmentation of neuroanatomy in MRI: a deep learning approach

Cited by 4 publications

References 29 publications

Synthetic FLAIR as a Substitute for FLAIR Sequence in Acute Ischemic Stroke

Synthetic FLAIR as a Substitute for FLAIR Sequence in Acute Ischemic Stroke

Semantic Image Segmentation: Two Decades of Research

Learn to Ignore: Domain Adaptation for Multi-Site MRI Analysis

Contact Info

Product

Resources

About