Objective: To explore the diagnostic value of quantitative CT parameters of radiomics in benign pulmonary nodules and malignant tumors.Methods: We retrospectively analyzed data from 60 patients with pathologically confirmed pulmonary nodules at our hospital between June 2023 and June 2025, dividing them into a malignant tumor group (36 cases) and a benign nodule group (24 cases). All patients underwent plain chest CT. Nine quantitative CT parameters were extracted using Pyradiomics software, including morphological (sphericity), first-order statistical (mean, standard deviation, skewness, kurtosis), and texture features (energy, contrast, correlation, entropy). Concurrently, senior radiologist subjective scores and serum CEA/CYFRA21-1 levels were recorded as reference standards. Diagnostic performance was evaluated using ROC curves. Results: Compared to the benign group, the malignant group exhibited significantly higher skewness, kurtosis, and entropy values, along with significantly lower sphericity (P<0.05). Among individual parameters, entropy (E4) demonstrated the highest diagnostic performance with an AUC of 0.910 (95% CI: 0.831–0.989), sensitivity of 91.67%, and specificity of 83.33%. The AUC for kurtosis (A5) was 0.892. The AUC for entropy (E4) was significantly superior to that of serum markers CEA (0.725) and CYFRA21-1 (0.751) (P<0.05). The parallel combination of kurtosis (A5) and entropy (E4) increased sensitivity to 94.44%. Conclusion: Quantitative CT parameters, particularly the texture feature “entropy” and the first-order statistic “kurtosis,” demonstrate high diagnostic value in distinguishing benign from malignant pulmonary nodules. They outperform certain conventional serum tumor markers and represent a promising, objective, non-invasive adjunct diagnostic tool.