Perceptual Influence: Improving the Perceptual Loss Design for Low-Dose CT Enhancement

2025-09-26

AI Summary

Current perceptual loss designs for low-dose CT (LDCT) image enhancement lack systematic guidance, leading to suboptimal clinical structural fidelity. Method: We propose “perceptual impact”—a novel metric—and establish the first perceptual loss analysis framework tailored to medical imaging. Our approach quantitatively evaluates the influence of feature hierarchy, pretraining dataset (ImageNet vs. medical-domain pretraining), and perceptual weight on reconstruction quality—without modifying network architecture. It integrates multi-level features, adopts medical-prior-aligned pretrained encoders, and determines optimal perceptual weights via statistical analysis. Contribution/Results: On LDCT denoising, our method achieves superior trade-offs between noise suppression and anatomical structure preservation. It outperforms pixel-based losses and state-of-the-art perceptual losses in both quantitative metrics (PSNR/SSIM) and radiologist assessments. The implementation is publicly available.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Web Mining and Content Analysis: Large pretrained models with web data
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets

Abstract

Perceptual losses have emerged as powerful tools for training networks to enhance Low-Dose Computed Tomography (LDCT) images, offering an alternative to traditional pixel-wise losses such as Mean Squared Error, which often lead to over-smoothed reconstructions and loss of clinically relevant details in LDCT images. The perceptual losses operate in a latent feature space defined by a pretrained encoder and aim to preserve semantic content by comparing high-level features rather than raw pixel values. However, the design of perceptual losses involves critical yet underexplored decisions, including the feature representation level, the dataset used to pretrain the encoder, and the relative importance assigned to the perceptual component during optimization. In this work, we introduce the concept of perceptual influence (a metric that quantifies the relative contribution of the perceptual loss term to the total loss) and propose a principled framework to assess the impact of the loss design choices on the model training performance. Through systematic experimentation, we show that the widely used configurations in the literature to set up a perceptual loss underperform compared to better-designed alternatives. Our findings show that better perceptual loss designs lead to significant improvements in noise reduction and structural fidelity of reconstructed CT images, without requiring any changes to the network architecture. We also provide objective guidelines, supported by statistical analysis, to inform the effective use of perceptual losses in LDCT denoising. Our source code is available at https://github.com/vngabriel/perceptual-influence.

Problem

• Improving perceptual loss design for low-dose CT image enhancement
• Analyzing feature representation and encoder pretraining choices
• Quantifying perceptual loss contribution to optimize denoising performance

Innovation

• Proposed perceptual influence metric for loss analysis
• Systematically evaluated perceptual loss design configurations
• Provided guidelines for perceptual loss in CT denoising