Using Out-of-the-Box Frameworks for Contrastive Unpaired Image Translation for Vestibular Schwannoma and Cochlea Segmentation: An approach for the crossMoDA Challenge

2021/10/02 by Choi, Jae Won
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #electronic engineering #information engineering

paper · doi:10.48550/arxiv.2110.01607

Abstract

The purpose of this study is to apply and evaluate out-of-the-box deep learning frameworks for the crossMoDA challenge. We use the CUT model, a model for unpaired image-to-image translation based on patchwise contrastive learning and adversarial learning, for domain adaptation from contrast-enhanced T1 MR to high-resolution T2 MR. As data augmentation, we generate additional images with vestibular schwannomas with lower signal intensity. For the segmentation task, we use the nnU-Net framework. Our final submission achieved mean Dice scores of 0.8299 in the validation phase and 0.8253 in the test phase. Our method ranked 3rd in the crossMoDA challenge.

Related

• Unsupervised Cross-Modality Domain Adaptation for Segmenting Vestibular Schwannoma and Cochlea with Data Augmentation and Model Ensemble
• Multi-view Cross-Modality MR Image Translation for Vestibular Schwannoma and Cochlea Segmentation
• MS-MT: Multi-Scale Mean Teacher with Contrastive Unpaired Translation for Cross-Modality Vestibular Schwannoma and Cochlea Segmentation
• Self-Training Based Unsupervised Cross-Modality Domain Adaptation for Vestibular Schwannoma and Cochlea Segmentation
• Weakly Unsupervised Domain Adaptation for Vestibular Schwannoma Segmentation