Multimodal machine learning predicts cerebral microhemorrhage risk in Alzheimer’s disease cohort

Comprehensive multimodal model achieved AUC 0.86 for baseline cerebral microhemorrhage detection in ADNI; minimal clinical-only model still reached AUC 0.72, supporting resource-limited screening.

• A multimodal ML framework integrating clinical, fluid biomarker, and imaging data was developed on 813 ADNI participants to predict presence, incidence, and stability of cerebral microhemorrhages.
• The full model detected baseline microhemorrhages with AUC 0.86; the minimal model using only demographics and clinical history yielded an AUC of 0.72.
• Longitudinal models predicted time-to-incidence (R²=0.67) and identified a transient vascular instability phenotype linked to hepatic factors.

Automated summary

RadPigeon summaries are original and for information only. They are not clinical advice.