CT-DegradBench: A Physics-Informed Benchmark for CT Degradation Detection and Severity Estimation

2026-05-14

AI Summary

This work addresses the challenge of evaluating diverse artifacts in computed tomography (CT) images, a task hindered by the limited clinical relevance and inconsistent applicability of existing assessment methods across multiple degradation types. To bridge this gap, the authors introduce CT-DegradBench, the first physics-aware benchmark encompassing a wide spectrum of CT artifacts, along with SeSpeCT, a training-free framework that jointly estimates both the type and severity of image degradations. By integrating semantic priors from medical vision-language models, frequency-domain features, and radiology-informed textual prompts, SeSpeCT achieves superior performance over current baselines in both single-type and mixed-degradation scenarios. The method enables accurate, interpretable, and unsupervised analysis of CT image quality without requiring annotated training data.

Abstract

Computed tomography (CT) images are frequently degraded by acquisition artifacts, including noise, blur, streaking, aliasing, and metal artifacts. Yet CT enhancement is still largely evaluated using image quality metrics with limited perceptual and clinical validity, while existing datasets remain focused on isolated restoration tasks, hindering unified benchmarking across diverse degradation types. We present CT-DegradBench, a dataset and benchmark for CT degradation detection and severity estimation under controlled single- and mixed-artifact settings. CT-DegradBench enables systematic evaluation across multiple degradation families and severity levels within a common experimental framework. We further propose SeSpeCT (Semantic-Spectral CT degradation estimation), a framework that combines semantic priors from medical vision-language models with complementary frequency-domain cues for artifact analysis. SeSpeCT constructs a training-free semantic quality axis in the multimodal embedding space using radiology-informed text prompts, without task-specific fine-tuning, and combines it with spectral features that capture degradation-specific frequency patterns. The resulting representation enables joint prediction of artifact type and severity. Experimental results show that SeSpeCT consistently outperforms the evaluated baselines under both single- and mixed-degradation settings. The framework is available at https://github.com/yousranb/CT-DEGRADBENCH.

Problem

• CT degradation
• artifact detection
• severity estimation
• image quality assessment
• benchmarking

Innovation

• physics-informed benchmark
• degradation detection
• severity estimation
• vision-language model
• frequency-domain analysis