AMLP: Adjustable Masking Lesion Patches for Self-Supervised Medical Image Segmentation

TL;DR

AMLP introduces an adaptive self-supervised framework for medical image segmentation that dynamically selects lesion patches, refines patch categorization, and gradually increases masking ratio to improve lesion detail capture and segmentation accuracy.

Contribution

The paper proposes AMLP, a novel self-supervised framework with adaptive masking, lesion-focused patch selection, and loss functions to enhance medical image segmentation performance.

Findings

Outperforms state-of-the-art self-supervised methods on medical datasets.

Effectively captures detailed lesion information for accurate segmentation.

Addresses challenges of fixed masking ratios in medical image modeling.

Abstract

Self-supervised masked image modeling (MIM) methods have shown promising performances on analyzing natural images. However, directly applying such methods to medical image segmentation tasks still cannot achieve satisfactory results. The challenges arise from the facts that (i) medical images are inherently more complex compared to natural images, and the subjects in medical images often exhibit more distinct contour features; (ii) moreover, the conventional high and fixed masking ratio in MIM is likely to mask the background, limiting the scope of learnable information. To address these problems, we propose a new self-supervised medical image segmentation framework, called Adjustable Masking Lesion Patches (AMLP), which employs Masked Patch Selection~(MPS) strategy to identify patches with high probabilities of containing lesions to help model achieve precise lesion reconstruction. To improve the categorization of patches in MPS, we further introduce Relative Reconstruction Loss (RRL) to better learn hard-to-reconstruct lesion patches. Then, Category Consistency Loss (CCL) is proposed to refine patch categorization based on reconstruction difficulty, enhancing difference between lesions and backgrounds. Moreover, an Adjustable Masking Ratio (AMR) strategy is proposed to gradually increase the masking ratio over training to expand~the scope of learnable mutual information. Extensive~experiments on two medical segmentation datasets demonstrate the superior performances of the proposed AMLP w.r.t. the SOTA self-supervised methods; the results prove that AMLP effectively addresses the challenges of applying masked modeling to medical images and capturing accurate lesion details that are crucial for segmentation tasks.