Research progress on risk stratification warning models for early hematoma expansion and poor outcome in intracerebral hemorrhage: A narrative review

This review explores recent advances in risk stratification systems for early hematoma expansion (HE), focusing on predictive models using CTA, NCCT, radiomics, and artificial intelligence to enhance accuracy and clinical utility. The study analyzes literature on CTA spot signs, NCCT imaging features, and radiomics combined with machine learning for hematoma expansion prediction. It compares traditional biomarkers with artificial intelligence-driven models.

A narrative review was performed to synthesize the available literature and explore the predictive value of CTA spot signs, NCCT imaging features, and radiomics combined with machine learning for hematoma expansion and outcomes.

Integrating multimodal imaging and artificial intelligence significantly enhances HE prediction. Future work should focus on multicenter validation and standardized models to optimize personalized treatment and improve intracerebral hemorrhage outcomes.

Intelligent risk stratification systems, incorporating multimodal imaging and artificial intelligence, enhance hematoma expansion prediction and clinical utility. Future research could prioritize multicenter validation, standardization of models, and the development of personalized treatment strategies to ultimately improve outcomes for ICH patients.