2020/09/27 by Haoliang Li, YuFei Wang, Yufei Wang +11 · 82 citations
Computer Science · Engineering · Mathematics · #AI in cancer detection #Artificial intelligence #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Dependency (UML) #Domain (mathematical analysis) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #FOS: Electrical engineering #Generalization #Image and Video Processing (eess.IV) #Machine Learning (cs.LG) #Mathematical analysis #Mathematics #Medical Image Segmentation Techniques #Pattern recognition (psychology) #Perspective (graphical) #Regularization (linguistics) #cs.CV #cs.LG #eess.IV #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2009.12829
published in arXiv (Cornell University) (Cornell University) · Accepted by NeurIPS, 2020
openalex publication_date 2020/09/27 · arxiv created 2020/10/29 · arxiv updated 2020/10/30 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Recently, we have witnessed great progress in the field of medical imaging classification by adopting deep neural networks. However, the recent advanced models still require accessing sufficiently large and representative datasets for training, which is often unfeasible in clinically realistic environments. When trained on limited datasets, the deep neural network is lack of generalization capability, as the trained deep neural network on data within a certain distribution (e.g. the data captured by a certain device vendor or patient population) may not be able to generalize to the data with another distribution. In this paper, we introduce a simple but effective approach to improve the generalization capability of deep neural networks in the field of medical imaging classification. Motivated by the observation that the domain variability of the medical images is to some extent compact, we propose to learn a representative feature space through variational encoding with a novel linear-dependency regularization term to capture the shareable information among medical data collected from different domains. As a result, the trained neural network is expected to equip with better generalization capability to the "unseen" medical data. Experimental results on two challenging medical imaging classification tasks indicate that our method can achieve better cross-domain generalization capability compared with state-of-the-art baselines.