CURRENT NEURAL NETWORKS DEMONSTRATE POTENTIAL IN AUTOMATED CERVICAL VERTEBRAL MATURATION STAGE CLASSIFICATION BASED...

Area of Science:

• Orthodontics
• Radiology
• Artificial Intelligence

Background:

• Cervical vertebrae maturation (CVM) assessment is crucial for orthodontic treatment planning.
• Traditional methods for CVM assessment can be subjective and time-consuming.
• Advancements in artificial intelligence (AI) offer potential for automated and objective CVM classification.

Purpose of the Study:

• To systematically review the literature on the application of neural networks for classifying cervical vertebrae maturation.
• To evaluate the accuracy and efficacy of different neural network models in CVM assessment.
• To identify trends and challenges in using AI for CVM analysis.

Main Methods:

• A systematic review of studies published on neural networks for CVM classification was conducted.
• Databases such as PubMed, Scopus, and Web of Science were searched using relevant keywords.
• Studies were screened based on predefined inclusion and exclusion criteria.

Main Results:

• Several studies demonstrated the feasibility of using neural networks for CVM classification.
• Convolutional Neural Networks (CNNs) showed promising results in accurately identifying maturation stages.
• The performance of AI models varied depending on the dataset size, network architecture, and training parameters.

Conclusions:

• Neural networks show significant potential as a tool for objective and efficient cervical vertebrae maturation classification.
• Further research is needed to standardize methodologies and validate AI models in diverse clinical settings.
• AI-driven CVM assessment could enhance the precision and predictability of orthodontic treatment outcomes.