DoFE: Domain-oriented Feature Embedding for Generalizable Fundus Image Segmentation on Unseen Datasets

Abstract

Deep convolutional neural networks have significantly boosted the performance of fundus image segmentation when test datasets have the same distribution as the training datasets. However, in clinical practice, medical images often exhibit variations in appearance for various reasons, e.g., different scanner vendors and image quality. These distribution discrepancies could lead the deep networks to over-fit on the training datasets and lack generalization ability on the unseen test datasets. To alleviate this issue, we present a novel Domain-oriented Feature Embedding (DoFE) framework to improve the generalization ability of CNNs on unseen target domains by exploring the knowledge from multiple source domains. Our DoFE framework dynamically enriches the image features with additional domain prior knowledge learned from multi-source domains to make the semantic features more discriminative. Specifically, we introduce a Domain Knowledge Pool to learn and memorize the prior information extracted from multi-source domains. Then the original image features are augmented with domain-oriented aggregated features, which are induced from the knowledge pool based on the similarity between the input image and multi-source domain images. We further design a novel domain code prediction branch to infer this similarity and employ an attention-guided mechanism to dynamically combine the aggregated features with the semantic features. We comprehensively evaluate our DoFE framework on two fundus image segmentation tasks, including the optic cup and disc segmentation and vessel segmentation. Our DoFE framework generates satisfying segmentation results on unseen datasets and surpasses other domain generalization and network regularization methods.

arXiv commentAccepted at IEEE-TMI 2020

References in corpus (16)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Prototypical Networks for Few-shot Learning
• mixup: Beyond Empirical Risk Minimization
• CE-Net: Context Encoder Network for 2D Medical Image Segmentation
• REFUGE Challenge: A Unified Framework for Evaluating Automated Methods for Glaucoma Assessment from Fundus Photographs
• Joint Optic Disc and Cup Segmentation Based on Multi-label Deep Network and Polar Transformation
• Manifold Mixup: Better Representations by Interpolating Hidden States
• Optic Disc and Cup Segmentation Methods for Glaucoma Detection with Modification of U-Net Convolutional Neural Network
• Disc-aware Ensemble Network for Glaucoma Screening from Fundus Image
• Patch-based Output Space Adversarial Learning for Joint Optic Disc and Cup Segmentation
• Task Driven Generative Modeling for Unsupervised Domain Adaptation: Application to X-ray Image Segmentation
• Domain and Geometry Agnostic CNNs for Left Atrium Segmentation in 3D Ultrasound
• When Unseen Domain Generalization is Unnecessary? Rethinking Data Augmentation
• Towards continual learning in medical imaging

Cited by in corpus (13)

• Domain Generalization: A Survey
• Towards Out-Of-Distribution Generalization: A Survey
• Generalist Vision Foundation Models for Medical Imaging: A Case Study of Segment Anything Model on Zero-Shot Medical Segmentation
• AADG: Automatic Augmentation for Domain Generalization on Retinal Image Segmentation
• Domain Generalization for Medical Image Analysis: A Review
• FedDG: Federated Domain Generalization on Medical Image Segmentation via Episodic Learning in Continuous Frequency Space
• MLN-net: A multi-source medical image segmentation method for clustered microcalcifications using multiple layer normalization
• A Dual-Task Synergy-Driven Generalization Framework for Pancreatic Cancer Segmentation in CT Scans
• A New Perspective to Boost Performance Fairness for Medical Federated Learning
• Semantic-Rearrangement-Based Multi-Level Alignment for Domain Generalized Segmentation
• BCMDA: Bidirectional Correlation Maps Domain Adaptation for Mixed Domain Semi-Supervised Medical Image Segmentation
• FunduSegmenter: Leveraging the RETFound Foundation Model for Joint Optic Disc and Optic Cup Segmentation in Retinal Fundus Images
• Domain Composition and Attention for Unseen-Domain Generalizable Medical Image Segmentation