AI model fuses MRI, pathology, and text for brain tumor classification

The DS@GT ARC team has developed a multimodal model for brain tumor subtype classification, combining MRI embeddings, histopathology embeddings, and radiology reports. Their system utilizes task-specific gates and explores different report encoders, including RadBERT and Llama 3.1 8B-Instruct. The model achieved a mean macro-F1 score of 0.801 on the MEDIQA-CORE 2026 Task 1, surpassing the baseline and securing second place among verified submissions. However, the system's performance is highly dependent on the availability of histopathology data. AI

IMPACT This research demonstrates advanced multimodal AI integration for complex medical diagnosis, potentially improving accuracy and speed in clinical settings.

RANK_REASON Research paper detailing a novel AI model for a specific medical task. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• DS@GT ARC
• GitHub
• ImageCLEFmed
• Llama 3.1 8B-Instruct
• MEDIQA-CORE-Task-1 2026
• NeuroVFM
• Prov-GigaPath

• paper
• model release
• product

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