Random Style Transfer Based Domain Generalization Networks Integrating Shape and Spatial Information

Li, Lei; Zimmer, Veronika A.; Ding, Wangbin; Wu, Fuping; Huang, Liqin; Schnabel, Julia A.; Zhuang, Xiahai

doi:10.1007/978-3-030-68107-4_21

Cited by 14 publications

(8 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our method was tested on 320 scans containing 98,810 slices which were accessible through the M&Ms challenge's online resource. Segmentation results were evaluated [17] 0.939 0.886 U-Net [20] 0.927 0.877 DRUNET [28] 0.922 0.857 SDNet [19] 0.889 0.835 U-Net [30] 0.896 0.772 U-Net [21] 0.797 0.716 Proposed EAD 0.873 0.770 only on the 5,273 annotated slices. Annotations were only available for the slices at the end of diastole and end of systole.…”

Section: Resultsmentioning

confidence: 99%

“…Additionally, domain adversarial training of neural networks (DANN) and domain unlearning (DU) were used to counter the problem of image heterogeneity. Li et al [21] augmented the domain using a random style transfer technique to dampen the impact of cross-domain scans. Zhang et al [22] used histogram matching to augment the domain, in addition to this, they also used unlabelled slices for label propagation to fine-tune the results.…”

Section: Prior Workmentioning

confidence: 99%

See 1 more Smart Citation

Automatic Segmentation of Left Ventricle in Cardiac Magnetic Resonance Images

Chhabra¹,

Gagan²,

Kumar³

2022

Preprint

View full text Add to dashboard Cite

Segmentation of the left ventricle in cardiac magnetic resonance imaging MRI scans enables cardiologists to calculate the volume of the left ventricle and subsequently its ejection fraction. The ejection fraction is a measurement that expresses the percentage of blood leaving the heart with each contraction. Cardiologists often use ejection fraction to determine one's cardiac function. We propose multiscale template matching technique for detection and an elliptical active disc for automated segmentation of the left ventricle in MR images. The elliptical active disc optimizes the local energy function with respect to its five free parameters which define the disc. Gradient descent is used to minimize the energy function along with Green's theorem to optimize the computation expenses. We report validations on 320 scans containing 5,273 annotated slices which are publicly available through the Multi-Centre, Multi-Vendor, and Multi-Disease Cardiac Segmentation (M&Ms) Challenge. We achieved successful localization of the left ventricle in 89.63% of the cases and a Dice coefficient of 0.873 on diastole slices and 0.770 on systole slices. The proposed technique is based on traditional image processing techniques with a performance on par with the deep learning techniques.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Prior Workmentioning

confidence: 99%

Automatic Segmentation of Left Ventricle in Cardiac Magnetic Resonance Images

Chhabra¹,

Gagan²,

Kumar³

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…Maier et al [120] provide an introduction to deep learning in medical imaging. Li et al [121] use style transfer to generate images from different vendors, such generated images enable better machine learning. Fu et al [122] and Haskins et al [123] provide an overview of machine learning in registrations.…”

Section: Machine Learningmentioning

confidence: 99%

Evaluating Registrations of Serial Sections With Distortions of the Ground Truths

Lobachev¹,

Funatomi

Pfaffenroth

et al. 2021

IEEE Access

View full text Add to dashboard Cite

“…The model and training process do not change. The second (MID-Net [11]) and third method (RST-Net [9,18]) are state-of-the-art methods employing different approaches to achieve DG. In MID-Net, domain-invariant features are extracted by mutual information based disentanglement in the latent space, while in RST-Net available domains are augmented via pseudo-novel domains.…”

Section: Domain Generalization Modelsmentioning

confidence: 99%

“…2 schemes for LA segmentation of multi-center LGE MRIs. The schemes include histogram matching (HM) [10], mutual information based disentangled (MID) representation [11], and random style transfer (RST) [9,18].…”

Section: Introductionmentioning

confidence: 99%

AtrialGeneral: Domain Generalization for Left Atrial Segmentation of Multi-Center LGE MRIs

Li¹,

Zimmer²,

Schnabel³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Left atrial (LA) segmentation from late gadolinium enhanced magnetic resonance imaging (LGE MRI) is a crucial step needed for planning the treatment of atrial fibrillation. However, automatic LA segmentation from LGE MRI is still challenging, due to the poor image quality, high variability in LA shapes, and unclear LA boundary. Though deep learning-based methods can provide promising LA segmentation results, they often generalize poorly to unseen domains, such as data from different scanners and/or sites. In this work, we collect 140 LGE MRIs from different centers with different levels of image quality. To evaluate the domain generalization ability of models on the LA segmentation task, we employ four commonly used semantic segmentation networks for the LA segmentation from multi-center LGE MRIs. Besides, we investigate three domain generalization strategies, i.e., histogram matching, mutual information based disentangled representation, and random style transfer, where a simple histogram matching is proved to be most effective.

show abstract

Random Style Transfer Based Domain Generalization Networks Integrating Shape and Spatial Information

Cited by 14 publications

References 12 publications

Automatic Segmentation of Left Ventricle in Cardiac Magnetic Resonance Images

Automatic Segmentation of Left Ventricle in Cardiac Magnetic Resonance Images

Evaluating Registrations of Serial Sections With Distortions of the Ground Truths

AtrialGeneral: Domain Generalization for Left Atrial Segmentation of Multi-Center LGE MRIs

Contact Info

Product

Resources

About