Artificial intelligence (AI) uses in stereotactic radiosurgery (SRS): diagnosis with brain metastasis (BM) - A systematic review.
J Clin NeurosciJul 20, 2026 (epub)
Shiv P Desai, Yusuke S Hori, Fred C Lam et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review evaluated the use of artificial intelligence (AI) in diagnosing brain metastases (BM) during stereotactic radiosurgery (SRS). It included 18 studies published between 2015 and 2025, focusing on AI's ability to differentiate radiation necrosis from true tumor progression. The highest-performing AI models achieved AUCs of 0.85-0.91, indicating strong diagnostic performance.
Key findings
- AI models showed AUCs ranging from 0.71 to 0.94 for differentiating radiation necrosis from tumor progression.
- The highest-performing models reached AUCs of 0.85-0.91 through multimodal integration of imaging and clinical data.
- Contrast-enhanced T1-weighted MRI was the dominant imaging input across studies.
- AI consistently outperformed expert neuroradiologist reads on matched cases.
Why it matters
The integration of AI in the diagnostic pathway for brain metastases could enhance accuracy in distinguishing between radiation effects and tumor progression, potentially reducing unnecessary procedures and improving patient management.