Knowledge-Guided Large Language Model for Automatic Pediatric Dental Record Understanding and Safe Antibiotic Recommendation
- 1School of Automation and Electrical Engineering, University of Jinan, Jinan, China
- 2Jilin University, Changchun, China
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.
Related
- Knowledge and misconceptions about antibiotic use and antibiotic resistance among secondary school students in Riyadh, Saudi Arabia: implications for school- based antibiotic resistance education
- MRSA Knowledge, Infection-Prevention Practices, and Antibiotic Use Without Clinician Consultation Among Hajj Pilgrims Arriving in Madinah, Saudi Arabia: A Cross-Sectional Study †
- Research priorities and knowledge gaps in understanding hepatobiliary disease in cystic fibrosis.
- Knowledge of antibiotic resistance and prevalence of antibiotic usage among urban residents of Kochi, India: a community-based cross-sectional study
- Beyond deficit and coexistence: Modeling the knowledge–conspiracy–mistrust configuration in public understanding of science
BibTeX
@article{han2026knowledgeguided,
title = {{Knowledge-Guided Large Language Model for Automatic Pediatric Dental Record Understanding and Safe Antibiotic Recommendation}},
author = {Zihan Han and Zhitao Shu and Caifeng Li},
journal = {Fundamental Scientific Reports in Multidisciplinary Areas},
year = {2026},
month = jun,
doi = {10.66238/fsrma91},
url = {https://doi.org/10.66238/fsrma91},
note = {CC BY},
}
RIS
TY - JOUR
TI - Knowledge-Guided Large Language Model for Automatic Pediatric Dental Record Understanding and Safe Antibiotic Recommendation
AU - Zihan Han
AU - Zhitao Shu
AU - Caifeng Li
PY - 2026
DA - 2026/06/10
JO - Fundamental Scientific Reports in Multidisciplinary Areas
DO - 10.66238/fsrma91
UR - https://doi.org/10.66238/fsrma91
AB - 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.
SN - 3134-7509
SN - 3134-7517
ER -