New MEDMKG knowledge graph unifies medical imaging and clinical data for AI

Researchers have developed MEDMKG, a novel multimodal knowledge graph designed to integrate visual and textual medical information. This graph combines data from MIMIC-CXR, a database of chest radiographs and reports, with the structured clinical knowledge of the Unified Medical Language System (UMLS). MEDMKG employs a multi-stage construction process, including concept extraction and relationship modeling using both rule-based tools and large language models, and introduces a new filtering algorithm called Neighbor-aware Filtering (NaF) for improved graph quality. Evaluations demonstrate that MEDMKG enhances performance on downstream medical tasks and provides a robust foundation for multimodal knowledge integration in medical AI. AI

IMPACT Enhances medical AI by integrating visual and textual data, potentially improving diagnostic accuracy and treatment planning.

RANK_REASON The cluster describes a new academic paper introducing a novel multimodal knowledge graph for medical AI. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• Hugging Face
• MEDMKG
• MIMIC-CXR
• Neighbor-aware Filtering (NaF)
• ScienceCast
• Unified Medical Language System
• Xiaochen Wang

• paper
• infra

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