Mixed dentition space analysis is a critical component of early orthodontic diagnosis; however, conventional manual digital methods are time-consuming and technique-sensitive. This study aims to develop and validate an artificial intelligence (AI)–based tool capable of performing mixed dentition space analysis using standardized intra-oral photographs in children aged 6–10 years.

This development and validation study will include children in the mixed dentition phase attending the Department of Dentistry, AIIMS Raipur. High-quality standardized frontal and occlusal intra-oral photographs will be captured using a mirrorless camera with calibration aids. Images will be manually annotated by trained pediatric dentists to mark mesial and distal tooth landmarks. Data augmentation techniques will be applied to improve model robustness. The AI model will be trained and validated using a deep learning framework with five-fold cross-validation.

The AI tool will automatically detect teeth, identify landmarks, calculate arch space, and perform space analysis using established methods such as Tanaka-Johnston, Moyers, and Boston University analyses. The diagnostic accuracy of the AI tool will be compared with manual digital space analysis performed on intra-oral photographs and intra-oral scans. Accuracy will be assessed using mean absolute difference, reliability using kappa statistics, and efficiency by time comparison.

This study aims to establish a reliable and efficient AI-based alternative for early orthodontic space assessment using intra-oral photographs.