Knowledge-Guided Large Language Model for Automatic Pediatric Dental Record Understanding and Safe Antibiotic Recommendation

TL;DR

This paper introduces a knowledge-guided large language model that enhances pediatric dental record understanding and antibiotic safety recommendations by integrating knowledge graphs, retrieval mechanisms, and safety validation, significantly improving accuracy and safety over baseline models.

Contribution

The study presents a novel KG-LLM framework combining knowledge graphs, retrieval-augmented generation, and safety validation for improved dental record interpretation and antibiotic recommendation.

Findings

Improves record-understanding F1 score to 0.914

Increases drug-dose-duration accuracy to 0.782

Reduces unsafe antibiotic suggestions by 50%

Abstract

Accurate interpretation of pediatric dental clinical records and safe antibiotic prescribing remain persistent challenges in dental informatics. Traditional rule-based clinical decision support systems struggle with unstructured dental narratives, incomplete radiographic descriptions, and complex safety constraints. To address these limitations, this study proposes a Knowledge-Guided Large Language Model (KG-LLM) that integrates a pediatric dental knowledge graph, retrieval-augmented generation (RAG), and a multi-stage safety validation pipeline for evidence-grounded antibiotic recommendation. The framework first employs a clinical NER/RE module to extract structured entities and relations from dental notes and radiology reports. Relevant guidelines, drug-safety rules, and analogous historical cases are subsequently retrieved from the knowledge graph and supplied to the LLM for diagnostic summarization and dose-drug-duration prediction. Safety assurance is achieved through a dual-layer validation mechanism combining deterministic rule checking with a learned classifier for detecting allergies, contraindications, and dosing errors. Experiments on 32,000 de-identified pediatric dental visit records demonstrate the effectiveness of the proposed approach. Compared with a domain-adapted Llama-2 clinical baseline, KG-LLM improves record-understanding performance (F1: 0.914 vs. 0.867), drug-dose-duration accuracy (Top-1: 0.782 vs. 0.716), and reduces unsafe antibiotic suggestions by 50%. Additional evaluation across summary quality, recommendation accuracy, and global safety scores further confirms the robustness of the system. Ablation analyses indicate that the knowledge graph, RAG, and safety modules each contribute substantially to clinical reliability and interpretability.