Value of Endobronchial Ultrasound-Derived Radiomics in Differentiating Pulmonary Sarcoidosis from Mediastinal Lymph Node Tuberculosis

INTRODUCTION: The aim of the study was to explore the feasibility of an ultrasound radiomics machine-learning model based on endobronchial ultrasound (EBUS) for differentiating pulmonary sarcoidosis from mediastinal lymph node tuberculosis. METHODS: Clinical characteristics and ultrasound images from 100 patients diagnosed with pulmonary sarcoidosis and 70 diagnosed with mediastinal lymph node tuberculosis were collected. Statistical analysis was performed to compare clinical features, such as age, sex, smoking history, and lymph node size, between the two groups. The least absolute shrinkage and selection operator was used to analyze the radiomics features extracted from EBUS-based ultrasound images. A support vector machine (SVM) algorithm was applied to establish an EBUS-based radiomics model, and statistically significant clinical features were incorporated to optimize the model. A total of 170 lymph nodes were randomly divided into the training (n = 119) and validation (n = 51) groups, and the diagnostic performance of the model was assessed using receiver operating characteristic (ROC) curves, area under the curve (AUC), accuracy, sensitivity, and specificity. RESULTS: Seven stable radiomics features with nonzero coefficients and four clinical features were selected as model inputs. The SVM model demonstrated great performance in both groups. In the training group, the ROC AUC of the SVM model was 0.909 (95% CI: 0.897-0.922), with 88.2% accuracy, 82.4% sensitivity, and 92.6% specificity. In the validation group, the ROC AUC was 0.917 (95% CI: 0.901-0.934), with 80.4% accuracy, 68.4% sensitivity, and 87.5% specificity. CONCLUSION: The SVM model based on EBUS radiomics and clinical data shows preliminary feasibility for differentiating pulmonary sarcoidosis from mediastinal lymph node tuberculosis. It may be used as an auxiliary tool in clinical practice and provides a potentially useful approach for the early diagnosis of these conditions.