Artificial intelligence-assisted endobronchial ultrasound for differentiating between benign and malignant thoracic lymph nodes: a meta-analysis

Abstract

Background: Endobronchial ultrasound (EBUS) is a widely used imaging modality for evaluating thoracic lymph nodes (LNs), particularly in the staging of lung cancer. Artificial intelligence (AI)-assisted EBUS has emerged as a promising tool to enhance diagnostic accuracy. However, its effectiveness in differentiating benign from malignant thoracic LNs remains uncertain. This meta-analysis aimed to evaluate the diagnostic performance of AI-assisted EBUS compared to the pathological reference standards.

Methods: A systematic search was conducted across PubMed, Embase, and Web of Science for studies assessing AI-assisted EBUS in differentiating benign and malignant thoracic LNs. The reference standard included pathological confirmation via EBUS-guided transbronchial needle aspiration, surgical resection, or other histological/cytological validation methods. Sensitivity, specificity, diagnostic likelihood ratios, and diagnostic odds ratio (OR) were pooled using a random-effects model. The area under the receiver operating characteristic curve (AUROC) was summarized to evaluate diagnostic accuracy. Subgroup analyses were conducted by study design, lymph node location, and AI model type.

Results: Twelve studies with a total of 6,090 thoracic LNs were included. AI-assisted EBUS showed a pooled sensitivity of 0.75 (95% confidence interval [CI]: 0.60-0.86, I² = 97%) and specificity of 0.88 (95% CI: 0.83-0.92, I² = 96%). The positive and negative likelihood ratios were 6.34 (95% CI: 4.41-9.08) and 0.28 (95% CI: 0.17-0.47), respectively. The pooled diagnostic OR was 22.38 (95% CI: 11.03-45.38), and the AUROC was 0.90 (95% CI: 0.88-0.93). The subgroup analysis showed higher sensitivity but lower specificity in retrospective studies compared to prospective ones (sensitivity: 0.87 vs. 0.42; specificity: 0.80 vs. 0.93; both p < 0.001). No significant differences were found by lymph node location or AI model type.

Conclusion: AI-assisted EBUS shows promise in differentiating benign from malignant thoracic LNs, particularly those with high specificity. However, substantial heterogeneity and moderate sensitivity highlight the need for cautious interpretation and further validation.

Systematic review registration: PROSPERO CRD42025637964.

Keywords: Artificial intelligence; Endobronchial ultrasound; Malignancy; Meta-analysis; Thoracic lymph node.

Fig. 1

Flowchart representing the process of study screening and selection for the meta-analysis. This includes identification through database searching, removal of duplicates, title/abstract screening, full-text review, and final inclusion

Fig. 2

Forest plots showing the pooled diagnostic performance of AI-assisted EBUS for differentiating benign and malignant thoracic lymph nodes across the 16 datasets. (A) Forest plot of pooled sensitivity with 95% confidence intervals; (B) Forest plot of pooled specificity with 95% confidence intervals. AI, artificial intelligence; EBUS, endobronchial ultrasound; LNs, lymph nodes; CI, confidence interval

Fig. 3

SROC curve for the diagnostic accuracy of AI-assisted EBUS in differentiating malignant from benign thoracic lymph nodes. The plot includes the summary operating point (pooled sensitivity and specificity), 95% confidence ellipse, and 95% prediction ellipse. The AUC quantifies overall diagnostic performance. AI, artificial intelligence; EBUS, endobronchial ultrasound; LNs, lymph nodes; SENS, sensitivity; SPEC, specificity; AUC, area under the curve; SROC, summary receiver operating characteristic

Fig. 4

Deek’s funnel plot for assessing publication bias among the included studies. The X-axis represents the inverse root of the effective sample size, and the Y-axis displays the log of the diagnostic odds ratio. The symmetry of the plot is used to evaluate the presence of bias. A non-significant p-value (p = 0.39) suggests low risk of publication bias. AI, artificial intelligence; EBUS, endobronchial ultrasound; LNs, lymph nodes; OR, odds ratio; ESS, effective sample size