Foundation model decodes pediatric headaches from brain scans

Researchers have developed a foundation model approach using rs-fMRI data to classify pediatric headaches. The NeuroSTORM model, when fine-tuned, demonstrated a notable ability to distinguish between children with and without headaches, achieving an AUROC of 0.82. While it showed promise in differentiating chronic migraine from other headache subtypes, its performance in classifying less common headache types was less precise. This study suggests that foundation models can effectively extract relevant brain activity patterns from limited rs-fMRI data for diagnostic purposes. AI

IMPACT Demonstrates potential for foundation models in medical diagnosis using neuroimaging data.

RANK_REASON Academic paper detailing a new foundation model approach for medical classification. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Guilherme Seidyo Imai Aldeia
• Hugging Face
• NeuroSTORM
• pediatric headache
• ScienceCast

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →