Diagnostic performance of photon-counting detector CT versus energy-integrating detector CT for colorectal polyp...

To compare the diagnostic performance of photon-counting detector CT (PCD-CT) and energy-integrating detector CT (EID-CT) for colorectal polyp detection using a phantom model with histopathologically verified polyps.

A phantom containing 84 histopathologically confirmed colorectal polyps from resected specimens underwent sequential PCD-CT and EID-CT scanning. Polyps were categorized by size as ≤ 5 mm, 6-9 mm, or ≥ 10 mm. Two blinded abdominal radiologists evaluated image quality (5-point Likert scale), quantitative metrics (signal-to-noise ratio [SNR], contrast-to-noise ratio [CNR]), detection sensitivity, and diagnostic confidence (3-point scale). Wilcoxon signed-rank and paired t-tests compared image quality parameters, McNemar's test compared detection rates, and inter-reader agreement was assessed using ICC and kappa statistics.

PCD-CT demonstrated significantly superior qualitative image quality compared to EID-CT, with higher scores for overall quality (5 [4, 5] vs. 3 [3, 4]), sharpness (5 [4, 5] vs. 3 [1, 4]), and lower noise levels (5 [4, 5] vs. 3 [3, 3]) (all p < 0.001).Quantitative analysis showed superior SNR and CNR across all polyp sizes (all p < 0.001), with the most pronounced improvement in ≤ 5 mm polyps (SNR: 143% increase, CNR: 149% increase). PCD-CT significantly improved detection rates for ≤ 5 mm (71.4% vs. 17.9%, p < 0.001) and 6-9 mm polyps (92.9% vs. 71.4%, p = 0.031). Reader confidence was also significantly higher with PCD-CT for ≤ 5 mm (p < 0.001) and 6-9 mm (p < 0.001) polyps, with substantial inter-reader agreement (kappa = 0.858; 95% CI:0.703-0.906).

PCD-CT provides significantly improved diagnostic accuracy for colorectal polyp detection compared to EID-CT, with substantial gains in small polyp sensitivity and overall image quality.