Habitat-based radiomics from contrast-enhanced CT and clinical data to predict lymph node metastasis in clinical N0 peripheral lung adenocarcinoma ≤ 3 cm

Abstract

This study aims to develop an integrated model combining habitat-based radiomics and clinical data to predict lymph node metastasis in patients with clinical N0 peripheral lung adenocarcinomas measuring ≤ 3 cm in diameter. We retrospectively analyzed 1132 patients with lung adenocarcinoma from two centers who underwent surgical resection with lymph node dissection and had preoperative computed tomography (CT) scans showing peripheral nodules ≤ 3 cm. Multivariable logistic regression was employed to identify independent risk factors for the clinical model. Radiomics and habitat models were constructed by extracting and analyzing radiomic features and habitat regions from contrast-enhanced CT images. Subsequently, a combined model was developed by integrating habitat-based radiomic features with clinical characteristics. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC). The habitat model exhibited promising predictive performance for lymph node metastasis, outperforming other standalone models with AUCs of 0.962, 0.865, and 0.853 in the training, validation, and external test cohorts, respectively. The combined model demonstrated superior discriminative ability, achieving the highest AUCs of 0.983, 0.950, and 0.877 for the training, validation, and external test cohorts, respectively. The integration of habitat-based radiomic features with clinical data offers a non-invasive approach to assess the risk of lymph node metastasis, potentially supporting clinicians in optimizing patient management decisions.

Keywords: Habitat imaging; Lymph node metastasis; Peripheral lung adenocarcinomas; Radiomics.

Fig. 1

Flowchart of the study subject selection. Center A, Guangxi Medical University Cancer Hospital; Center B, Affiliated Hospital of Youjiang Medical University for Nationalities.

Fig. 2

Overall workflow of the work.

Fig. 3

Generated habitat regions.

Fig. 4

(a) Assessment of clustering performance using the Calinski–Harabasz (CH) Index, Silhouette Coefficient (SC), and Davies–Bouldin (DB) Index across varying numbers of clusters. (b) Visualization of habitat features segmented into three clusters. CH Index: quantifies the ratio of between-cluster dispersion to within-cluster compactness, with higher values reflecting superior clustering quality. SC: Evaluates the compactness of each sample within its cluster relative to other clusters, with values approaching 1 indicating optimal clustering. DB Index: Evaluates the ratio of within-cluster scatter to between-cluster separation, with lower values signifying improved clustering performance.

Fig. 5

(a) Number and proportion of handcrafted features across categories. (b) Coefficients derived from 10-fold cross-validation for the habitat model. (c) Histogram illustrating intratumor heterogeneity based on the selected habitat features. (d) Mean Squared Error (MSE) obtained from ten-fold cross-validation.

Fig. 6

Performance evaluation of different models across cohorts. (a–c) Receiver Operating Characteristic (ROC) curves for the (a) training, (b) validation, and (c) external test cohorts. (d–f) Calibration curves for the (d) training, (e) validation, and (f) external test cohorts. (g–i) Delong test results for the (g) training, (h) validation, and (i) external test cohorts.

Fig. 7

Decision curve analysis (DCA) for the combined model in the (a) training, (b) validation, and (c) external test cohort.

Fig. 8

Nomogram for clinical use.