Assesment of Ethnic Bias in an Artificial Intelligence Based Orthodontic Diagnosis System

NCT07162753 | Recruiting DMC

• 1 other identifier: E-54022451-050.04-177270
• Study Type: observational
• Target: 5,000
• Locations: 1 country
• Sites: 1

Brief Summary

This study explores how artificial intelligence (AI) can be used in orthodontics, which is the area of dentistry that focuses on correcting jaw and bite problems. AI is a computer technology that can learn from large amounts of data and then make predictions or decisions. It is already being tested in medicine and dentistry to help doctors and dentists diagnose conditions. For this study, the AI system was trained using photographs and X-rays from patients in Turkey. The system learned to recognize specific orthodontic skeletal malocclusions. After the training stage, the AI was tested in two groups: one group included Turkish patients whose records were not used in training, and the other group included patients from different ethnic backgrounds who were treated at a clinic in Belgium. This design allows researchers to see if the AI works equally well for people of different backgrounds. Only photographs and X-rays taken before orthodontic treatment are used in the study, and all data are anonymized so that no personal information is shared. The images must meet certain quality standards. For example the head must be in natural position, with no beards, scars, or previous orthodontic treatment that might affect the image. Patients who do not meet these criteria are not included. The AI program analyzes the profile photographs, prepares them for evaluation by adjusting and standardizing the images, and then tries to decide each patient has which malocclusion. The results from Turkish patients and patients from other ethnic groups are compared to see if the system makes fair and accurate decisions for everyone. The purpose of this study is not to test a new treatment, but to understand how well AI can recognize orthodontic problems in different populations. This information is important because AI systems are increasingly being used in healthcare, and they need to be fair and accurate for all patients, not just those from one group. By participating, patients help researchers learn whether AI in orthodontics is reliable across diverse communities. This knowledge can guide future improvements in AI technology, ensuring that it supports orthodontists in providing safe, equal, and effective care for everyone.

Conditions

Class II Division 2 Malocclusion; Class III Malocclusion; Class I Malocclusion; Class II Div 1 Malocclusion

Keywords

Artificial Intelligence; Deep Learning; Ethnic Bias; Machine Learning; Orthodontics

Outcome Measures

Primary Outcomes (1)

• Accuracy

  Accuracy (proportion of correctly classified cases) will be compared between Turkish test set, and non-Turkish cohort from different ethnic backgrounds.

  At baseline (retrospective single time-point data analysis)

Secondary Outcomes (4)

• Precision (Positive Predictive Value)

  At baseline (retrospective single time-point data analysis)

• Recall

  At baseline (retrospective single time-point data analysis)

• False Positive Rate

  At baseline (retrospective single time-point data analysis)

• F1 score

  At baseline (retrospective single time-point data analysis)

Study Arms (2)

Turkish Citizen's Control Group (Hold-out)

The AI system was trained with data from 7,000 people. To test whether the AI has bias, the system should be evaluated both on a group of Turkish citizens who were not included in the training data and on patients from different ethnic backgrounds.

Individuals from different ethnic backgrounds

Test set, consisting of individuals from different ethnic backgrounds.

Eligibility Criteria

• Age: 4 Years+
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

One group consists of Turkish citizens who were not included in the AI training set. Another group consists of patients who visited the PEBS clinic in Brussels, Belgium, representing individuals from different ethnic backgrounds.

You may not qualify if:

• Blurry missing or non standard profile photographs
• Presence of beard mustache scars cleft lip palate or soft tissue deformities that may significantly affect facial appearance
• Previous orthodontic treatment history

Sponsors & Collaborators

• Bezmialem Vakif University: lead
• PEBS DENTAL CLINIC: collaborator

Study Sites (1)

Bezmialem Vakıf University

Istanbul, Fatih, 34020, Turkey (Türkiye)

RECRUITING

Related Publications (2)

• Allareddy V, Oubaidin M, Rampa S, Venugopalan SR, Elnagar MH, Yadav S, Lee MK. Call for algorithmic fairness to mitigate amplification of racial biases in artificial intelligence models used in orthodontics and craniofacial health. Orthod Craniofac Res. 2023 Dec;26 Suppl 1:124-130. doi: 10.1111/ocr.12721. Epub 2023 Oct 17.

  PMID: 37846615 BACKGROUND

• Kilic B, Ibrahim AH, Aksoy S, Sakman MC, Demircan GS, Onal-Suzek T. A family-centered orthodontic screening approach using a machine learning-based mobile application. J Dent Sci. 2024 Jan;19(1):186-195. doi: 10.1016/j.jds.2023.05.001. Epub 2023 May 17.

  PMID: 38303845 BACKGROUND

MeSH Terms

Conditions

Malocclusion, Angle Class I; Malocclusion, Angle Class II; Malocclusion, Angle Class III

Condition Hierarchy (Ancestors)

Malocclusion; Tooth Diseases; Stomatognathic Diseases

Study Officials

• Banu Kılıç, Doctor of Orthodontics

  Bezmialem Vakif University

  STUDY DIRECTOR

Central Study Contacts

Aslı Eker Davut, PhD Student

CONTACT

00905356173181; aslieker95@gmail.com

Banu Kılıç, Doctor

CONTACT

00905322432756; bkilic@bezmialem.edu.tr

Study Design

• Study Type: observational
• Observational Model: OTHER
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Dentist - PhD Student

Study Record Dates

• First Submitted: August 26, 2025
• First Posted: September 9, 2025
• Study Start: January 6, 2025
• Primary Completion (Estimated): June 15, 2026
• Study Completion (Estimated): August 11, 2026
• Last Updated: December 5, 2025

Record last verified: 2025-07

Data Sharing

• IPD Sharing: Will share
• Shared Documents: STUDY PROTOCOL, SAP, ICF, CSR
• Time Frame: Beginning 3 months and ending a year after the publication of results
• Access Criteria: The investigators can access to the corresponding author by e-mail.

Locations

TU (1)