Real-Time Artificial Intelligence Driven Intraoral Scanner for Craniofacial Growth Analysis

Sunny A.; Laxmikanth S.M.; Sharan P.; Chetan V.; Rueda M. · Proceedings - 2025 IEEE International Conference on Compute, Control, Network and Photonics, ICCCNP 2025 (2025) · DOI: 10.1109/icccnp63914.2025.11233504

Resumen

Orthodontic treatment plan for the use of cephalometric evaluations especially for craniofacial growth pattern analysis. Analyzing the mandibular growth pattern is crucial in deciding the mode of treatment to be planned. Previous studies show there exists a strong correlation between craniofacial growth pattern and palatal rugae patterns. This study developed and evaluated a non-invasive, artificial intelligence powered device which identifies craniofacial growth patterns through the analysis of palatal rugae, providing a safer, radiation-free alternative to traditional radiation diagnostics in orthodontics. Using a high-resolution camera integrated with a Raspberry Pi 4B, the device captures detailed images of the palatal rugae, which are then processed by a pre-trained InceptionV3-based convolutional neural network (CNN) to classify growth patterns as horizontal, vertical, or average. Designed for orthodontic diagnostics and treatment planning, the device offers real-time feedback on a thin film transistor (TFT) screen, helping orthodontists make precise, informed decisions. Its user-friendly interface and efficient processing capabilities enhance the accuracy of growth pattern analysis and treatment planning while demonstrating the potential of artificial intelligence in advancing non-invasive diagnostic methods.

Instituciones: Universidad Nacional Federico Villarreal