MixSearch: Searching for Domain Generalized Medical Image Segmentation Architectures
This addresses the challenge of poor generalization in medical image analysis due to small, domain-specific datasets, offering a solution for improved segmentation performance on unseen data.
The paper tackles the problem of limited generalization in medical image segmentation by proposing MixSearch, a method to search for domain-generalized U-shape architectures using a composited dataset from multiple tasks and domains, achieving state-of-the-art results across various datasets.
Considering the scarcity of medical data, most datasets in medical image analysis are an order of magnitude smaller than those of natural images. However, most Network Architecture Search (NAS) approaches in medical images focused on specific datasets and did not take into account the generalization ability of the learned architectures on unseen datasets as well as different domains. In this paper, we address this point by proposing to search for generalizable U-shape architectures on a composited dataset that mixes medical images from multiple segmentation tasks and domains creatively, which is named MixSearch. Specifically, we propose a novel approach to mix multiple small-scale datasets from multiple domains and segmentation tasks to produce a large-scale dataset. Then, a novel weaved encoder-decoder structure is designed to search for a generalized segmentation network in both cell-level and network-level. The network produced by the proposed MixSearch framework achieves state-of-the-art results compared with advanced encoder-decoder networks across various datasets.