Artificial Intelligence for Diagnosis, Risk Stratification, and Prognosis of Neuroblastoma - A Systematic Review and Meta-Analysis.
Daniel Rodrigo Serbena, Isabela Luiza Fraron Cieslack, Renan Cassiano Ratis et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review and meta-analysis evaluated the performance of artificial intelligence (AI) in diagnosing, risk stratifying, and prognosing neuroblastoma across 53 studies. Machine Learning Models (MLMs) showed an Area Under the Curve (AUC) of 0.87 for differential diagnosis, compared to 0.83 for radiologists, although this difference was not statistically significant. Hybrid Nomograms (HNs) achieved an AUC of 0.87 in risk stratification.
Key findings
- Machine Learning Models had an AUC of 0.87 for differential diagnosis compared to 0.83 for radiologists.
- Hybrid Nomograms achieved an AUC of 0.87 in risk stratification.
- AI-derived nomograms outperformed conventional prognostic markers with an AUC of 0.9.
- Only 33.9% of models reported calibration and 24.5% underwent external validation.
Why it matters
The findings highlight the potential of AI in enhancing neuroblastoma diagnosis and risk assessment. However, the lack of robust validation and calibration raises concerns about the readiness for clinical implementation, emphasizing the need for further research.