Development of an Artificial Intelligence-Based Clinical Image Model for Detection, Classification, and Management Recommendations of Anterior Gingival Recession

This study aims to develop and evaluate an artificial intelligence-based clinical image model for the detection, classification, and management recommendations of anterior gingival recession. The study will utilize clinical images of patients presenting with gingival recession to train and validate a machine learning model capable of accurately identifying and classifying the condition according to established clinical criteria. In addition, the model will provide preliminary treatment recommendations based on the severity and type of recession. This is a diagnostic and model-development study designed to support clinicians in improving the accuracy and consistency of diagnosis and treatment planning for gingival recession in the anterior region.

Study Overview

• Status: Completed
• Conditions: Gingival Recessions
• Intervention / Treatment: Diagnostic test: Artificial Intelligence-Based Clinical Image Analysis Model

Detailed Description

This study is designed to develop and validate an artificial intelligence (AI)-based clinical image analysis model for the detection, classification, and management recommendation of anterior gingival recession. Gingival recession is a common periodontal condition characterized by apical displacement of the gingival margin, which may lead to aesthetic concerns, dentinal hypersensitivity, and increased risk of root caries.

Clinical intraoral images of patients presenting with anterior gingival recession will be collected following standardized imaging protocols. The dataset will be used to train, validate, and test a machine learning model capable of identifying the presence of gingival recession and classifying its severity and/or type according to established periodontal classification systems.

The AI model will also be designed to generate preliminary management recommendations based on the detected class, supporting clinical decision-making. Model performance will be evaluated using standard metrics such as accuracy, sensitivity, specificity, precision, recall, and area under the receiver operating characteristic curve (AUC-ROC).

The study is observational in nature with a diagnostic and model-development component. All patient data will be anonymized to ensure confidentiality, and ethical approval will be obtained prior to data collection. The final output is intended to support clinicians in improving diagnostic consistency and treatment planning efficiency for anterior gingival recession.

• Study Type: Observational
• Enrollment (Actual): 149

Contacts and Locations

Study Locations

• Egypt
  • Cairo, Egypt
    • Faculty of Dental Medicine for Girls, Al-Azhar University

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

The study population will consist of adult patients presenting with gingival recession affecting anterior teeth and attending the outpatient clinics of the Faculty of Dental Medicine for Girls, Al-Azhar University. Participants with clinically visible anterior gingival recession and adequate clinical photographs suitable for image analysis will be included in the study.

Description

Inclusion Criteria:

• Patients aged 18 years or older
• Presence of at least one anterior tooth exhibiting gingival recession classified according to the Cairo classification system (RT1, RT2, or RT3). - The gingival margin must be clearly visible.
• High-quality images (good focus, lighting, and resolution) are required.
• Clinically visible and intact cementoenamel junction (CEJ).

Exclusion Criteria:

• Presence of cervical restorations or fixed prostheses that interfere with CEJ identification.
• Patients undergoing active orthodontic treatment.
• Pregnant individuals, due to hormonal changes affecting gingival tissues.
• Images with poor photographic quality.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Intervention / Treatment

Gingival Recession Patients; This group consists of patients presenting with anterior gingival recession. Clinical intraoral images will be collected from eligible participants and used for the development and validation of an artificial intelligence-based classification model. The dataset includes cases with varying degrees and types of gingival recession according to established clinical classification criteria. No therapeutic intervention will be performed as part of the study, and all images will be analyzed for diagnostic and classification purposes only.

Diagnostic test: Artificial Intelligence-Based Clinical Image Analysis Model; An artificial intelligence-based clinical image model will be developed and evaluated using standardized clinical photographs of anterior teeth presenting with gingival recession. The model will be trained to detect the presence of gingival recession, classify lesions according to the Cairo classification system (RT1, RT2, and RT3), and generate preliminary management recommendations based on the identified classification. The system's performance will be assessed by comparing its diagnostic and classification outputs with expert clinical assessments.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Sensitivity and specificity of the AI system in detecting gingival recession, compared to clinical probing measurements.	-Primary Outcome 1 Outcome Measure: Sensitivity and specificity of the AI system for detecting gingival recession compared with clinical probing measurements. Primary Outcome 2 Outcome Measure: Agreement between the AI system and expert clinicians in classifying gingival recession according to the Cairo classification, assessed using Cohen's kappa coefficient.	Through study completion, an average of 6 months

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
- Error in automated CEJ identification, compared to manual annotations.	Error in automated CEJ identification, compared to manual annotations.; Concordance rate between AI-generated treatment recommendations and those proposed by experienced periodontists.	Immediately after AI analysis of the clinical images

Collaborators and Investigators

• Sponsor: Al-Azhar University

Study record dates

Study Major Dates

• Study Start (Actual): June 15, 2025
• Primary Completion (Actual): January 15, 2026
• Study Completion (Actual): April 15, 2026

Study Registration Dates

• First Submitted: June 30, 2026
• First Submitted That Met QC Criteria: July 7, 2026
• First Posted (Actual): July 9, 2026

Study Record Updates

• Last Update Posted (Actual): July 9, 2026
• Last Update Submitted That Met QC Criteria: July 7, 2026
• Last Verified: July 1, 2026

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Periodontal Atrophy; Periodontal Diseases; Mouth Diseases; Stomatognathic Diseases; Gingival Diseases; Gingival Recession
• Other Study ID Numbers: OMPDR-108-1r

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No