Value of dual-layer spectral detector computed tomography in the diagnosis of benign/malignant solid solitary pulmonary nodules and establishment of a prediction model

This study aimed to investigate the role of spectral detector computed tomography (SDCT) quantitative parameters and their derived quantitative parameters combined with lesion morphological information in the differential diagnosis of solid SPNs. This retrospective study included basic clinical data and SDCT images of 132 patients with pathologically confirmed SPNs (102 and 30 patients in the malignant and benign groups, respectively). The morphological signs of SPNs were evaluated and the region of interest (ROI) was delineated from the lesion to extract and calculate the relevant SDCT quantitative parameters, and standardise the process. Differences in qualitative and quantitative parameters between the groups were statistically analysed. A receiver operating characteristic (ROC) curve was constructed to evaluate the efficacy of the corresponding parameters in the diagnosis of benign and malignant SPNs. Statistically significant clinical data, CT signs and SDCT quantitative parameters were analysed using multivariate logistic regression to determine the independent risk factors for predicting benign and malignant SPNs, and the best multi-parameter regression model was established. Inter-observer repeatability was assessed using the intraclass correlation coefficient (ICC) and Bland-Altman plots. Malignant SPNs differed from benign SPNs in terms of size, lesion morphology, short spicule sign, and vascular enrichment sign ( 0.05). The SDCT quantitative parameters and their derived quantitative parameters of malignant SPNs (SAR , SAR , Δ , Δ , CER , CER , NEF , NEF , λ, NIC, NZ ) were significantly higher than those of benign SPNs ( 0.05). In the subgroup analysis, most parameters could distinguish between benign and adenocarcinoma groups (SAR , SAR , Δ , Δ , CER , CER , NEF , NEF , λ, NIC, and NZ ), and between benign and squamous cell carcinoma groups (SAR , SAR70 , Δ , Δ , NEF , NEF , λ, and NIC). However, there were no significant differences between the parameters in the adenocarcinoma and squamous cell carcinoma groups. ROC curve analysis indicated that NIC, NEF , and NEF had higher diagnostic efficacy for differentiating benign and malignant SPNs (area under the curve [AUC]:0.869, 0.854, and 0.853, respectively), and NIC was the highest. Multivariate logistic regression analysis showed that size (OR=1.138, 95% CI 1.022-1.267, =0.019), Δ (OR=1.060, 95% CI 1.002-1.122, =0.043), and NIC (OR=7.758, 95% CI 1.966-30.612, =0.003) were independent risk facto