Accuracy of an AI-clinical Knowledge-based Hybrid System for Detecting Periodontitis in OPG Images

NCT06306677 | Recruiting

• 1 other identifier: SH9H-2023-T369-1
• Study Type: observational
• Target: 1,200
• Locations: 3 countries
• Sites: 3

Brief Summary

Periodontitis is highly prevalent and rarely detected and treated in the earlier stages of the disease. Orthopantomography (OPG) is the most frequently taken dental radiograph around the world, and its systematic screening may contribute to early detection of periodontitis and access to the needed level of care. The investigators' recent study initially developed an AI-clinical knowledge-based system for automatic periodontitis diagnosis and indicated good performance for differentiating stage II-IV periodontitis. This cross-sectional diagnostic study aims to compare the diagnostic accuracy of this AI-clinical knowledge-based hybrid system (Index test) with human experts (reference test) for differentiating stage II-IV periodontitis using the OPG images obtained from different 4 centers around the world.

Conditions

Periodontitis; Periodontal Diseases

Keywords

Periodontitis; Artificial Intelligence; Radiograph

Outcome Measures

Primary Outcomes (4)

• Sensitivity measure

  Sensitivity of the AI-clinical based hybrid system to correctly identify periodontitis cases labeled by panel of experts

  1 day

• Specificity measure

  Specificity of the AI-clinical based hybrid system to correctly identify periodontitis cases labeled by panel of experts

  1 day

• The area under the receiver operating characteristic curve (AUC) measure

  The area under the receiver operating characteristic curve (AUC) measure of AI-clinical based hybrid system to correctly identify periodontitis cases labeled by panel of experts

  1 day

• Diagnostic accuracy

  The overall diagnostic accuracy of the AI-clinical-based system will be calculated based on the fraction of true test results \[Accuracy = (true positives + true negatives) / (total)\] and compared with the panel of experts

  1 day

Secondary Outcomes (1)

• Assessment time

  1 day

Study Arms (1)

Subjects presenting for care at Hospitals

The group is formed by subjects reporting for care at one of the four participating hospitals (China, Hong Kong SAR, Italy) who required an OPG radiographs for their routine clinical care.

Diagnostic Test: AI-clincial-based hybrid system for radiographic image analysis

Interventions

AI-clincial-based hybrid system for radiographic image analysis DIAGNOSTIC_TEST

The Index test is a novel AI-clinical-based hybrid system for radiographic image analysis. Its diagnostic performance will be compared to the reference represented by a panel of experts.

Subjects presenting for care at Hospitals

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

1. Patients seeking care at Shanghai Ninth People's Hospital Pudong Clinics; 2. Patients seeking care at Shanghai Ninth People's Hospital South Clinics; 3. Patients seeking care at Prince Philip Dental Hospital in Hong Kong; 4. Patients seeking care at The University of Rome La Sapienza in Italy;

You may qualify if:

• Aged 18 and above
• Having taken the OPG image

You may not qualify if:

• Edentulous mouth

Sponsors & Collaborators

• Shanghai Ninth People's Hospital Affiliated to Shanghai Jiao Tong University: lead
• ShanghaiTech University: collaborator
• University of Roma La Sapienza: collaborator
• The University of Hong Kong: collaborator

Study Sites (3)

Shanghai Perio-Implant Innovation Center

Shanghai, Shanghai Municipality, 201206, China

RECRUITING

Prince Philip Dental Hospital

Hong Kong, Hong Kong

RECRUITING

La Sapienza Dental School

Roma, Italy

RECRUITING

Related Publications (1)

• Li Y, Cui Z, Mei L, Xie Y, Marini L, Pelekos G, Gu W, Yu X, Wu X, Wei X, Tao L, Deng K, Pilloni A, Shen D, Tonetti MS. A novel AI-powered radiographic analysis surpasses specialists in stage II-IV periodontitis detection: a multicenter diagnostic study. NPJ Digit Med. 2025 Nov 18;8(1):691. doi: 10.1038/s41746-025-02077-0.

  PMID: 41254181 DERIVED

MeSH Terms

Conditions

Periodontitis; Periodontal Diseases

Condition Hierarchy (Ancestors)

Mouth Diseases; Stomatognathic Diseases

Study Officials

• Maurizio Tonetti

  Shanghai Ninth People Hospital

  PRINCIPAL INVESTIGATOR

Central Study Contacts

Maurizio Tonetti

CONTACT

15000102368; maurizio.tonetti@ergoperio.eu

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: CROSS SECTIONAL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Professor

Study Record Dates

• First Submitted: March 4, 2024
• First Posted: March 12, 2024
• Study Start: March 12, 2024
• Primary Completion: August 1, 2024
• Study Completion: December 31, 2024
• Last Updated: March 13, 2024

Record last verified: 2024-03

Data Sharing

• IPD Sharing: Will not share

Locations

HK (1); CN (1); IT (1)