Research on Deep Learning-Based Intelligent Diagnosis and Treatment

NCT07551622 | Not Yet Recruiting

• 1 other identifier: 2023-XJKQIEC-043-002
• Study Type: observational
• Target: 2,000
• Locations: 0 countries
• Sites: N/A

Brief Summary

This study aims to develop and evaluate deep learning-based artificial intelligence models for craniomaxillofacial multi-modal imaging analysis and clinical decision support. Approximately 2,000 participants with craniomaxillofacial imaging data and related clinical information will be included. The imaging data may include two-dimensional facial photographs, cone-beam computed tomography images, and three-dimensional facial surface scans. The study will use artificial intelligence methods to analyze craniofacial images and identify clinically meaningful features related to facial morphology, skeletal or dental classification, anatomical landmarks, regional structures, and craniomaxillofacial abnormalities. The models will be developed for tasks such as image classification, anatomical landmark detection, image segmentation, abnormality recognition, and treatment-related decision support. The purpose of this study is to improve the accuracy, efficiency, and consistency of image-based assessment in dentistry, orthodontics, and oral and maxillofacial clinical practice. The artificial intelligence models developed in this study are intended to provide objective imaging analysis and decision-support information for health care providers. These models are designed to assist clinicians and will not replace professional diagnosis or individualized treatment planning by qualified clinicians. This research may benefit patients and families by supporting earlier and more accurate recognition of craniomaxillofacial conditions, improving communication about diagnosis and treatment options, and promoting more personalized oral health care. All clinical images and related information will be handled according to approved research procedures and privacy protection requirements.

Conditions

Malocclusion; Dentofacial Deformities; Craniofacial Morphology

Outcome Measures

Primary Outcomes (1)

• Diagnostic performance of artificial intelligence models for craniomaxillofacial imaging analysis

  The primary outcome is the diagnostic performance of the developed artificial intelligence models on the independent testing dataset. Performance will be evaluated using accuracy, precision, recall, F1-score, sensitivity, specificity, positive predictive value, negative predictive value, receiver operating characteristic curve analysis, and area under the receiver operating characteristic curve, as appropriate for the specific classification or diagnostic recognition task.

  At completion of model validation on the independent testing dataset, expected within 12 months after study initiation

Study Arms (1)

Craniomaxillofacial Imaging Cohort

Participants with available craniomaxillofacial imaging data and related clinical information obtained during routine dental, orthodontic, oral and maxillofacial clinical care.

Eligibility Criteria

• Age: 6 Years - 70 Years
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

The study population will include approximately 2,000 participants with craniomaxillofacial imaging data and related clinical information obtained during routine dental, orthodontic, oral and maxillofacial, or related clinical care. Eligible participants may have two-dimensional facial photographs, cone-beam computed tomography images, or three-dimensional facial surface scans available for artificial intelligence-based imaging analysis. Related clinical information may include demographic characteristics, clinical diagnosis, skeletal or dental classification, cephalometric measurements, treatment-related records, and expert assessment results. The study will use available clinical imaging data to develop and validate deep learning models for craniomaxillofacial image classification, segmentation, landmark detection, abnormality recognition, and treatment-related decision support.

You may qualify if:

• Participants with available craniomaxillofacial imaging data obtained during routine dental, orthodontic, oral and maxillofacial, or related clinical care.
• Participants with at least one eligible imaging modality, including two-dimensional facial photographs, cone-beam computed tomography images, or three-dimensional facial surface scans.
• Participants with related clinical information available for model development or validation, such as demographic information, clinical diagnosis, skeletal or dental classification, cephalometric measurements, treatment-related records, or expert assessment results.
• Imaging data of sufficient quality for artificial intelligence-based image analysis, annotation, segmentation, landmark detection, classification, or decision-support model development.

You may not qualify if:

• Participants with incomplete or unavailable key imaging data or clinical information required for the planned analysis.
• Images with severe artifacts, poor resolution, incorrect orientation, incomplete anatomical coverage, or other quality problems that prevent reliable analysis.
• Duplicate records or repeated imaging records that cannot be reliably linked to a unique participant.
• Participants whose data cannot be used according to institutional review board approval, consent requirements, or applicable privacy protection regulations.

Sponsors & Collaborators

• Xi'an Jiaotong University: lead

MeSH Terms

Conditions

Malocclusion; Dentofacial Deformities

Condition Hierarchy (Ancestors)

Tooth Diseases; Stomatognathic Diseases; Maxillofacial Abnormalities; Craniofacial Abnormalities; Musculoskeletal Abnormalities; Musculoskeletal Diseases; Stomatognathic System Abnormalities; Congenital Abnormalities; Congenital, Hereditary, and Neonatal Diseases and Abnormalities

Central Study Contacts

Yu xin Guo, Ph.D

CONTACT

+86 17782626326; guoyuxin@xjtu.edu.cn

Hui Yang, master candidate

CONTACT

+86 15859811830; yanghui1585981@163.com

Study Design

• Study Type: observational
• Observational Model: COHORT
• Time Perspective: RETROSPECTIVE
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Research Assistant Professor

Study Record Dates

• First Submitted: April 19, 2026
• First Posted: April 27, 2026
• Study Start: May 1, 2026
• Primary Completion (Estimated): December 31, 2027
• Study Completion (Estimated): December 31, 2029
• Last Updated: April 27, 2026

Record last verified: 2026-04

Data Sharing

• IPD Sharing: Will not share