Age group prediction at the 18-year-old threshold using radiographic images of the mandibular third molar of a Thai population: A comparison between human-based and deep learning-based methods

Status

Active, not recruiting

Phases

Unknown

Study type

Observational

Source

TCTR

Registry ID

TCTR20230519002

Enrollment

2000

Registered

2023-05-19

Start date

2012-01-01

Completion date

Unknown

Last updated

2026-08-03

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

Conditions

Dental age estimation using deep learning Dental age estimation Deep learning Mandibular third molar Forensic Odontology Demirjian method

Interventions

Patients who were under 18 years old,Patients who were over 18 years old

Health Services Research,Health Services Research

Under 18 years old,Over 18 years old

Eligibility

Sex/Gender

All

Age

14 Years to 23 Years

Inclusion criteria

Inclusion criteria: Patients with available data on the birth date and date of radiographic examination

Exclusion criteria

Exclusion criteria: 1. Patients whose radiographic images had poor quality, 2. Patients with missing or malaligned mandibular third molars (severe buccoversion or linguoversion), 3. Patients who had developmental anomalies, jawbone pathology, or syndromes that affected the dental development

Design outcomes

Primary

Measure	Time frame
Accuracy of age-group prediction After the model's training phase Percentage of accuracy	—

Secondary

Measure	Time frame
N/A N/A N/A	—

Countries

Thailand

Contacts

Public ContactWitsarut Upalananda

Faculty of Dentistry, Prince of Songkla University

witsarut.u@psu.ac.th6674287716

Outcome results

None listed

Source: TCTR (via WHO ICTRP) · Data processed: Aug 9, 2026