Collaborative Unsupervised Domain Adaptation for Medical Image Diagnosis

Abstract

Deep learning based medical image diagnosis has shown great potential in clinical medicine. However, it often suffers two major difficulties in real-world applications: 1) only limited labels are available for model training, due to expensive annotation costs over medical images; 2) labeled images may contain considerable label noise (e.g., mislabeling labels) due to diagnostic difficulties of diseases. To address these, we seek to exploit rich labeled data from relevant domains to help the learning in the target task via {Unsupervised Domain Adaptation} (UDA). Unlike most UDA methods that rely on clean labeled data or assume samples are equally transferable, we innovatively propose a Collaborative Unsupervised Domain Adaptation algorithm, which conducts transferability-aware adaptation and conquers label noise in a collaborative way. We theoretically analyze the generalization performance of the proposed method, and also empirically evaluate it on both medical and general images. Promising experimental results demonstrate the superiority and generalization of the proposed method.

arXiv commentIEEE Transactions on Image Processing

References in corpus (17)

• A Survey on Deep Learning in Medical Image Analysis
• Unsupervised Domain Adaptation by Backpropagation
• Deep Domain Confusion: Maximizing for Domain Invariance
• Conditional Adversarial Domain Adaptation
• PCANet: A Simple Deep Learning Baseline for Image Classification?
• Unsupervised Domain Adaptation with Residual Transfer Networks
• Training Deep Neural Networks on Noisy Labels with Bootstrapping
• MentorNet: Learning Data-Driven Curriculum for Very Deep Neural Networks on Corrupted Labels
• Domain Adaptation: Learning Bounds and Algorithms
• Asymmetric Tri-training for Unsupervised Domain Adaptation
• Dimensionality-Driven Learning with Noisy Labels
• Unsupervised Label Noise Modeling and Loss Correction
• Collaborative Unsupervised Domain Adaptation for Medical Image Diagnosis
• Co-regularized Alignment for Unsupervised Domain Adaptation
• COVID-DA: Deep Domain Adaptation from Typical Pneumonia to COVID-19
• Deep Multi-View Learning using Neuron-Wise Correlation-Maximizing Regularizers
• Multi-marginal Wasserstein GAN

Cited by in corpus (29)

• A Survey on Incorporating Domain Knowledge into Deep Learning for Medical Image Analysis
• Collaborative Unsupervised Domain Adaptation for Medical Image Diagnosis
• Self-Attentive Spatial Adaptive Normalization for Cross-Modality Domain Adaptation
• Explainable, Domain-Adaptive, and Federated Artificial Intelligence in Medicine
• Deep Long-Tailed Learning: A Survey
• Margin Preserving Self-paced Contrastive Learning Towards Domain Adaptation for Medical Image Segmentation
• COVID-DA: Deep Domain Adaptation from Typical Pneumonia to COVID-19
• Closed-loop Matters: Dual Regression Networks for Single Image Super-Resolution
• Unleashing the Power of Contrastive Self-Supervised Visual Models via Contrast-Regularized Fine-Tuning
• Cost-Sensitive Portfolio Selection via Deep Reinforcement Learning
• Distill-SODA: Distilling Self-Supervised Vision Transformer for Source-Free Open-Set Domain Adaptation in Computational Pathology
• Dense Regression Network for Video Grounding
• Source-free Domain Adaptation via Avatar Prototype Generation and Adaptation
• A Thorough Comparison Study on Adversarial Attacks and Defenses for Common Thorax Disease Classification in Chest X-rays
• Adaptive Hierarchical Dual Consistency for Semi-Supervised Left Atrium Segmentation on Cross-Domain Data
• Nearest Neighborhood-Based Deep Clustering for Source Data-absent Unsupervised Domain Adaptation
• Unsupervised Domain Adaptation for Vestibular Schwannoma and Cochlea Segmentation via Semi-supervised Learning and Label Fusion
• Disturbance-immune Weight Sharing for Neural Architecture Search
• Intelligent Home 3D: Automatic 3D-House Design from Linguistic Descriptions Only
• Online Adaptive Asymmetric Active Learning with Limited Budgets
• MoNetV2: Enhanced Motion Network for Freehand 3D Ultrasound Reconstruction
• Online Continual Adaptation with Active Self-Training
• Meta-learners for few-shot weakly-supervised optic disc and cup segmentation on fundus images
• Hard Class Rectification for Domain Adaptation
• TCSA-UDA: Text-Driven Cross-Semantic Alignment for Unsupervised Domain Adaptation in Medical Image Segmentation
• Retinal Image Segmentation with a Structure-Texture Demixing Network
• Towards Robust Cross-domain Image Understanding with Unsupervised Noise Removal
• An Empirical Framework for Domain Generalization in Clinical Settings
• Domain-Specific Bias Filtering for Single Labeled Domain Generalization