Lesion-Inspired Denoising Network: Connecting Medical Image Denoising and Lesion Detection

TL;DR

This paper introduces LIDnet, a novel framework that jointly optimizes medical image denoising and lesion detection by integrating detection feedback into the denoising process, significantly enhancing both tasks.

Contribution

The paper proposes a lesion-inspired denoising network with a novel ROI perceptual loss and collaborative training, connecting denoising and detection tasks for improved performance.

Findings

Significant improvement in denoising quality.

Enhanced lesion detection accuracy.

Effective on low-dose CT datasets.

Abstract

Deep learning has achieved notable performance in the denoising task of low-quality medical images and the detection task of lesions, respectively. However, existing low-quality medical image denoising approaches are disconnected from the detection task of lesions. Intuitively, the quality of denoised images will influence the lesion detection accuracy that in turn can be used to affect the denoising performance. To this end, we propose a play-and-plug medical image denoising framework, namely Lesion-Inspired Denoising Network (LIDnet), to collaboratively improve both denoising performance and detection accuracy of denoised medical images. Specifically, we propose to insert the feedback of downstream detection task into existing denoising framework by jointly learning a multi-loss objective. Instead of using perceptual loss calculated on the entire feature map, a novel region-of-interest (ROI) perceptual loss induced by the lesion detection task is proposed to further connect these two tasks. To achieve better optimization for overall framework, we propose a customized collaborative training strategy for LIDnet. On consideration of clinical usability and imaging characteristics, three low-dose CT images datasets are used to evaluate the effectiveness of the proposed LIDnet. Experiments show that, by equipping with LIDnet, both of the denoising and lesion detection performance of baseline methods can be significantly improved.