BoundarySeg:An Embarrassingly Simple Method To Boost Medical Image Segmentation Performance for Low Data Regimes
This addresses the challenge of costly and scarce annotated medical data for researchers and practitioners in medical imaging, offering an incremental improvement by leveraging existing annotations more effectively.
The paper tackles the problem of medical image segmentation in low data regimes by proposing BoundarySeg, a multi-task framework that uses boundary prediction as an auxiliary task to improve segmentation accuracy without needing unannotated data, achieving performance comparable to or exceeding state-of-the-art semi-supervised methods.
Obtaining large-scale medical data, annotated or unannotated, is challenging due to stringent privacy regulations and data protection policies. In addition, annotating medical images requires that domain experts manually delineate anatomical structures, making the process both time-consuming and costly. As a result, semi-supervised methods have gained popularity for reducing annotation costs. However, the performance of semi-supervised methods is heavily dependent on the availability of unannotated data, and their effectiveness declines when such data are scarce or absent. To overcome this limitation, we propose a simple, yet effective and computationally efficient approach for medical image segmentation that leverages only existing annotations. We propose BoundarySeg , a multi-task framework that incorporates organ boundary prediction as an auxiliary task to full organ segmentation, leveraging consistency between the two task predictions to provide additional supervision. This strategy improves segmentation accuracy, especially in low data regimes, allowing our method to achieve performance comparable to or exceeding state-of-the-art semi supervised approaches all without relying on unannotated data or increasing computational demands. Code will be released upon acceptance.