Dual-energy CT has shown promising results in determining tumor characteristics and treatment effectiveness through spectral data by assessing normalized iodine concentration (nIC), normalized effective atomic number (nZeff), normalized electron density (nED), and extracellular volume (ECV). This study explores the value of quantitative parameters in contrast-enhanced dual-layer spectral detector CT (SDCT) as a potential tool for detecting lymph node activity in lymphoma patients. A retrospective analysis of 55 lymphoma patients with 289 lymph nodes, assessed through ¹⁸FDG-PET/CT and the Deauville five-point scale, revealed significantly higher values of nIC, nZeff, nED, and ECV in active lymph nodes compared to inactive ones (p < 0.001). Generalized linear mixed models showed statistically significant fixed-effect parameters for nIC, nZeff, and ECV (p < 0.05). The area under the receiver operating characteristic curve (AUROC) values of nIC, nZeff, and ECV reached 0.822, 0.845, and 0.811 for diagnosing lymph node activity. In conclusion, the use of g nIC, nZeff, and ECV as alternative imaging biomarkers to PET/CT for identifying lymph node activity in lymphoma holds potential as a reliable diagnostic tool that can guide treatment decisions.

Background: Advanced Modeling Iterative Reconstruction (ADMIRE) algorithm has five intensity levels; it is important to study which algorithm is better for brain CT scanning. Objective: The aim of the study is to compare the influence of different strength levels of ADMIRE and traditional Filtered Back Projection (FBP) on image quality in brain CT scanning. Methods: 60 patients were retrospectively selected, and the data from each of these patients’ brains were reconstructed by four different reconstruction methods (FBP, ADMIRE1, ADMIRE3, and ADMIRE5). A five-point Likert Scale was implemented to evaluate the subjective image quality. Image noise, CT value of brain tissue , signal-to-noise ratio (SNR) of gray white matter, contrast-to-noise ratio (CNR), and beam hardening artifact index (AI) of the posterior fossa, were measured for evaluating the objective image quality. Finally, the differences between the subjective and objective evaluations were compared. Results: There were no statistical differences observed in CT values of gray matter and white matter between the four groups (all P >0.05). The image noise gradually decreased with the increase of ADMIRE algorithm level. The AI exhibited no statistical difference between the four groups (F =0.793, P =0.499), but it tended to decrease slightly with the increase of ADMIRE algorithm level. Compared to other groups (all p <0.001), the ADMIRE5 group demonstrated the best objective image quality. Nevertheless, the highest subjective score was observed in the ADMIRE3 group, which exhibited significant differences with other images (all P <0.001). Conclusion: ADMIRE algorithm can clearly improve image quality, but it cannot significantly improve the linear sclerosis artifacts in the posterior cranial fossa. Based on the subjective evaluation of image quality, ADMIRE3 algorithm is recommended in brain CT scanning.