A New Perspective on ADHD Research: Knowledge Graph Construction with LLMs and Network Based Insights
Hakan T. Otal, Stephen V. Faraone, M. Abdullah Canbaz
TL;DR
This work tackles ADHD's heterogeneous etiology by building a multimodal knowledge graph that fuses literature, clinical data, and expert-curated content via LLM-driven semantic extraction. The authors construct and refine a semantically rich graph using contextual proximity and redundancy reduction (via embeddings and DBSCAN), yielding a network of $N=2347$ nodes and $E=8655$ edges. Network analysis reveals a sparse but modular structure with key core concepts up to $k=16$, identifying central health conditions and cognitive outcomes, and distinct communities around comorbidities and neuroimaging/genetics. The resulting Graph-RAG framework supports accurate, context-aware information retrieval for ADHD research and clinical practice, while highlighting areas for data expansion and methodological refinement to enhance interpretability and decision support.
Abstract
Attention-Deficit/Hyperactivity Disorder (ADHD) is a challenging disorder to study due to its complex symptomatology and diverse contributing factors. To explore how we can gain deeper insights on this topic, we performed a network analysis on a comprehensive knowledge graph (KG) of ADHD, constructed by integrating scientific literature and clinical data with the help of cutting-edge large language models. The analysis, including k-core techniques, identified critical nodes and relationships that are central to understanding the disorder. Building on these findings, we curated a knowledge graph that is usable in a context-aware chatbot (Graph-RAG) with Large Language Models (LLMs), enabling accurate and informed interactions. Our knowledge graph not only advances the understanding of ADHD but also provides a powerful tool for research and clinical applications.