REHRSeg: Unleashing the Power of Self-Supervised Super-Resolution for Resource-Efficient 3D MRI Segmentation

TL;DR

REHRSeg introduces a resource-efficient framework that leverages self-supervised super-resolution to enable high-resolution 3D MRI segmentation using only low-resolution images, reducing the need for extensive high-res data.

Contribution

The paper presents a novel self-supervised super-resolution approach with uncertainty awareness and structural knowledge distillation for efficient 3D MRI segmentation.

Findings

Achieves high-quality HR segmentation from LR images.

Significantly improves baseline LR segmentation performance.

Reduces dependency on extensive high-resolution annotated data.

Abstract

High-resolution (HR) 3D magnetic resonance imaging (MRI) can provide detailed anatomical structural information, enabling precise segmentation of regions of interest for various medical image analysis tasks. Due to the high demands of acquisition device, collection of HR images with their annotations is always impractical in clinical scenarios. Consequently, segmentation results based on low-resolution (LR) images with large slice thickness are often unsatisfactory for subsequent tasks. In this paper, we propose a novel Resource-Efficient High-Resolution Segmentation framework (REHRSeg) to address the above-mentioned challenges in real-world applications, which can achieve HR segmentation while only employing the LR images as input. REHRSeg is designed to leverage self-supervised super-resolution (self-SR) to provide pseudo supervision, therefore the relatively easier-to-acquire LR annotated images generated by 2D scanning protocols can be directly used for model training. The main contribution to ensure the effectiveness in self-SR for enhancing segmentation is three-fold: (1) We mitigate the data scarcity problem in the medical field by using pseudo-data for training the segmentation model. (2) We design an uncertainty-aware super-resolution (UASR) head in self-SR to raise the awareness of segmentation uncertainty as commonly appeared on the ROI boundaries. (3) We align the spatial features for self-SR and segmentation through structural knowledge distillation to enable a better capture of region correlations. Experimental results demonstrate that REHRSeg achieves high-quality HR segmentation without intensive supervision, while also significantly improving the baseline performance for LR segmentation.