Banca de QUALIFICAÇÃO: LUANNY DE BRITO AVELINO CASSIANO

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : LUANNY DE BRITO AVELINO CASSIANO
DATE: 15/10/2024
TIME: 08:00
LOCAL: Online: https://meet.google.com/ixs-kgev-ays
TITLE:

USE OF ARTIFICIAL INTELLIGENCE FOR CALCULATING THE LENGTH OF MINI-IMPLANTS USED IN THE MARPE TECHNIQUE ON COMPUTED TOMOGRAPHY

KEY WORDS:

Artificial Intelligence; Palatal Expansion Technique; Orthodontic anchorage procedures; Cone-Beam Computed Tomography.

PAGES: 32
BIG AREA: Ciências da Saúde
AREA: Odontologia
SUBÁREA: Ortodontia
SUMMARY:

Introduction: For miniscrew-assisted rapid palatal expansion (MARPE), selecting miniscrews with effective lengths that ensure good primary stability through bicortical engagement is crucial. This requires manual measurement of the palatal thickness in maxillary cone-beam computed tomography (CBCT) scans, which is time-consuming and operator-dependent. In this context, automating these measurements using artificial intelligence (AI) presents a promising approach. Objective: To evaluate the efficacy of AI for segmenting the bone and mucosa of the palate in maxillary CBCT scans to determine the effective length of miniscrews. Materials and Methods: This is an observational, cross-sectional, descriptive, and inferential study. With a sample of 280 CBCT scans, the segmentation of the palatal mucosa and bone will be performed using the segment editor tool in the 3DSlicer software to train a neural network and obtain a palatal segmentation model. Appropriate metrics for evaluating model accuracy will be applied, such as the Dice Similarity Coefficient (DSC), Average Symmetric Surface Distance (ASSD), Intersection over Union (IoU), Mean Thickness Percentage Error (MTPE), Sensitivity, and Positive Predictive Value. A web application will then be developed with a palatal thickness measurement algorithm (Pixel Traversal Method) combined with a summation function in Python to determine the effective length of the miniscrews. A validation phase will compare the model's results with manual measurements performed by an orthodontist. Expected Results: The proposed model is expected to provide precise and reliable measurements for the miniscrew length, and the resulting application should offer speed, efficiency, and safety for the procedure.

COMMITTEE MEMBERS:
Presidente - 1009536 - SERGEI GODEIRO FERNANDES RABELO CALDAS
Interno - 1378325 - ARTHUR CESAR DE MEDEIROS ALVES
Externo ao Programa - 2885532 - IVANOVITCH MEDEIROS DANTAS DA SILVA - UFRN