Integrating CT image features and quantitative dual-energy CT parameters for predicting cervical lymph nodes metastasis from laryngeal and hypopharyngeal squamous cell carcinoma
10.3969/j.issn.1002-1671.2025.10.007
- VernacularTitle:联合CT图像特征与双能CT定量参数对喉及下咽鳞状细胞癌颈部淋巴结转移的预测价值
- Author:
Yujun HU
1
;
Guoyi SU
;
Hao HU
;
Haibing CHEN
;
Xi CHEN
;
Xiaoquan XU
;
Feiyun WU
Author Information
1. 南京医科大学第一附属医院放射科,江苏 南京 210029
- Publication Type:Journal Article
- Keywords:
laryngeal and hypopharyngeal squamous cell carcinoma;
lymph nodes metastasis;
dual-energy computed tomography;
prediction
- From:
Journal of Practical Radiology
2025;41(10):1624-1628
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the value of integrating CT image features and quantitative dual-energy computed tomography(DECT)parameters in predicting cervical lymph nodes metastasis from laryngeal and hypopharyngeal squamous cell carcinoma(LHSCC).Methods The clinical and imaging data of 99 patients with LHSCC confirmed by pathology were retrospectively analyzed.All patients were divided into metastatic group(41 cases)and non-metastatic group(58 cases).The CT image features,including location,size and depth,were analyzed,respectively.The quantitative DECT parameters in the arterial and venous phases including iodine concentration(IC)and normalized iodine concentration(NIC)were measured.The rank sum test or independent-samples t-test were used to compare the difference of CT image features and quantitative DECT parameters between the two groups.The multivariate logistic regression analysis was used to build the models based on CT image features(image feature model)and combination of CT image features and quantitative DECT parameters(combined model).The receiver operating characteristic(ROC)curve was performed to analyze and compare the difference of predictive efficiency between the two groups.Results There were significant differences in tumor location between the non-metastatic group and the metastatic group(χ2=21.736,P<0.001).Size(33.20 mm vs 24.95 mm,P<0.001),depth(21.10 mm vs 13.15 mm,P<0.001)and NIC in the arterial phase(0.18 vs 0.14,P<0.001)in the metastatic group were significantly higher than those in the non-metastatic group.The area under the curve(AUC),sensitivity,specificity,positive predictive value,negative predictive value,accuracy of the combined model were 0.851,75.6%,82.8%,58.5%,87.9%and 75.8%for predicting cervical lymph nodes metastasis.The AUC,sensitivity,specificity,positive predictive value,negative predictive value,accuracy of the image feature model were 0.792,95.1%,56.9%,53.7%,81.0%and 69.7%,respectively.The prediction performance of the combined model was better than that of the image feature model(Z=-2.028,P=0.043).Conclusion Integrating CT image features and quantitative DECT parameters has important value for predicting cervical lymph nodes metastasis from LHSCC.
