Artificial Intelligence-Driven MRI for Cervical Nodal Metastasis Detection in Oral Squamous Cell Carcinoma: A Hierarchical Meta-Analysis of Diagnostic Accuracy.
José Evando da Silva-Filho, Renata Roque Ribeiro, André Wescley Oliveira de Aguiar et al.
✦ AI-curated · Sources linked
In 30 seconds
This meta-analysis evaluated the diagnostic accuracy of AI-driven MRI for detecting cervical lymph node metastases in oral squamous cell carcinoma (OSCC) across 12 studies. The analysis included 548 participants, revealing a pooled sensitivity of 0.72 and specificity of 0.79, indicating moderate diagnostic performance. The study design was a hierarchical meta-analysis.
Key findings
- Pooled sensitivity of AI-assisted MRI was 0.72 (95% CI: 0.62-0.80).
- Pooled specificity was 0.79 (95% CI: 0.73-0.83).
- The area under the curve (AUC) was 0.82.
- The diagnostic odds ratio was 9.42.
Why it matters
Understanding the diagnostic performance of AI-assisted MRI can help clinicians assess its potential role in improving the detection of cervical lymph node metastases in patients with OSCC, which is crucial for treatment planning.
What to keep in mind
Evidence certainty was low, and heterogeneity was mainly related to threshold effects.