Problem
Research questions and friction points this paper is trying to address.
Photon-counting CT
Energy-integrating CT
Image quality enhancement
Clinical translation
Degradation modeling
Distilling Photon-Counting CT into Routine Chest CT through Clinically Validated Degradation Modeling
🤖 AI Summary
This work addresses the limited clinical adoption of photon-counting computed tomography (PCCT), despite its superior image quality, by proposing SUMI—a novel method that leverages a radiologist-validated, clinically realistic degradation model to translate high-quality PCCT images into realistic conventional CT images and inversely learns the enhancement process to synthesize PCCT-level images from standard CT scans. Requiring no large-scale paired data, SUMI combines autoencoder pretraining with a latent diffusion model to establish a generalizable CT feature representation and introduces a large-scale, publicly available annotated dataset. Experimental results demonstrate that SUMI improves SSIM by 15% and PSNR by 20% on external data, significantly enhances radiologist-assigned image quality scores, and boosts lesion detection sensitivity and F1 score by 15% and 10%, respectively.
📝 Abstract
Photon-counting CT (PCCT) provides superior image quality with higher spatial resolution and lower noise compared to conventional energy-integrating CT (EICT), but its limited clinical availability restricts large-scale research and clinical deployment. To bridge this gap, we propose SUMI, a simulated degradation-to-enhancement method that learns to reverse realistic acquisition artifacts in low-quality EICT by leveraging high-quality PCCT as reference. Our central insight is to explicitly model realistic acquisition degradations, transforming PCCT into clinically plausible lower-quality counterparts and learning to invert this process. The simulated degradations were validated for clinical realism by board-certified radiologists, enabling faithful supervision without requiring paired acquisitions at scale. As outcomes of this technical contribution, we: (1) train a latent diffusion model on 1,046 PCCTs, using an autoencoder first pre-trained on both these PCCTs and 405,379 EICTs from 145 hospitals to extract general CT latent features that we release for reuse in other generative medical imaging tasks; (2) construct a large-scale dataset of over 17,316 publicly available EICTs enhanced to PCCT-like quality, with radiologist-validated voxel-wise annotations of airway trees, arteries, veins, lungs, and lobes; and (3) demonstrate substantial improvements: across external data, SUMI outperforms state-of-the-art image translation methods by 15% in SSIM and 20% in PSNR, improves radiologist-rated clinical utility in reader studies, and enhances downstream top-ranking lesion detection performance, increasing sensitivity by up to 15% and F1 score by up to 10%. Our results suggest that emerging imaging advances can be systematically distilled into routine EICT using limited high-quality scans as reference.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Photon-counting CT
Degradation modeling
Latent diffusion model
Image enhancement
Medical image synthesis
🔎 Similar Papers
No similar papers found.
