DeepATLAS: One-Shot Localization for Biomedical Data
This addresses localization tasks for biomedical imaging, offering a one-shot approach that reduces the need for extensive labeled data, though it is incremental as it builds on existing self-supervised and few-shot methods.
The paper tackles the problem of localization in biomedical data by introducing DeepATLAS, a foundational model that achieves high one-shot segmentation performance on over 50 anatomic structures across four external test sets, matching or exceeding standard supervised models.
This paper introduces the DeepATLAS foundational model for localization tasks in the domain of high-dimensional biomedical data. Upon convergence of the proposed self-supervised objective, a pretrained model maps an input to an anatomically-consistent embedding from which any point or set of points (e.g., boxes or segmentations) may be identified in a one-shot or few-shot approach. As a representative benchmark, a DeepATLAS model pretrained on a comprehensive cohort of 51,000+ unlabeled 3D computed tomography exams yields high one-shot segmentation performance on over 50 anatomic structures across four different external test sets, either matching or exceeding the performance of a standard supervised learning model. Further improvements in accuracy can be achieved by adding a small amount of labeled data using either a semisupervised or more conventional fine-tuning strategy.