Skip to content

Development of an Decision-Making Expert-System for Dental Defect Restoration based on Multi-Modal Large Language Models

Development of an Decision-Making Expert-System for Dental Defect Restoration based on Multi-Modal Large Language Models

Status
Active, not recruiting
Phases
Early Phase 1
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2600119100
Enrollment
Unknown
Registered
2026-02-16
Start date
2026-02-17
Completion date
Unknown
Last updated
2026-03-02

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Tooth defect

Interventions

Gold Standard:An expert consensus that comprehensively considers multiple factors such as the amount of tooth defect, pulp condition, periodontal condition, and the value of tooth restoration and rete
Index test:A large language model that integrates expert experience and is competent for clinical assessment of dental defects and analysis and decision-making of transparent restoration types

Sponsors

Hospital of Stomatology, SunYat-sen University
Lead Sponsor

Eligibility

Sex/Gender
All
Age
13 Years to 80 Years

Inclusion criteria

Inclusion criteria: 1.The adjacent teeth of the patient's defective tooth are healthy; 2.Male or female, aged 13–80 years; 3.Meets the diagnostic criteria outlined in the Guidelines for the Diagnosis and Treatment of Dental Defects, Dentition Defects, and Missing Teeth Restoration issued by the National Health Commission in 2022; 4.CBCT clearly shows the presence of dental defects or cracks in the affected tooth; 5.The affected tooth ultimately undergoes (or does not undergo) dental defect restoration (e.g., full crown restoration, post-core crown restoration);

Exclusion criteria

Exclusion criteria: 1. Patients with partial or complete tooth loss where the remaining teeth may serve as abutments for restoration; 2. Pregnancy or lactation; 3. Incomplete or substandard imaging data based on general empirical clinical criteria.

Design outcomes

Primary

MeasureTime frame
Accuracy of AI-selected restoration type;

Secondary

MeasureTime frame
Cross-annotation agreement;Model performance (AUC, F1);General clinical characteristic indicators (defect amount of affected tooth, resistance of remaining dental tissue, retention shape of remaining dental tissue, pulp condition, periodontal tissue condition, patient age);Self-reflection Rate Score;

Countries

China

Contacts

Public ContactGuo Jiawen

Hospital of Stomatology, SunYat-sen University

guojw28@mail.sysu.edu.cn+86 136 3023 8526

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Mar 14, 2026