This retrospective validity study evaluates the accuracy of artificial intelligence (AI) in determining gender from mandibular morphometric linear measurements. The study utilizes pre-existing Cone Beam Computed Tomography (CBCT) scans of adult Egyptian dental patients. After automatic segmentation of the mandible from these scans, a radiologist will manually perform measurements from certain anatomical points. These measurements will be the reference standard for the AI models.
A three-dimensional deep learning model will be developed to perform two tasks:
1. To identify the anatomical points and make the specified linear measurements from these points on the segmented mandibles 2. To accurately predict gender based on these measurements. (Main Objective) The primary objective of this study is to evaluate the accuracy of machine learning algorithms in gender identification from linear morphometric measurements of the mandible. The known gender from patient records will serve as the reference standard.
This study will assess the reliability of AI as an objective tool for gender determination for forensic purposes.
Scans belonging to patients above or at the age of 18 years (≥ 18 years)
Scans that image the area of interest clearly (mandible) with a craniofacial or maxillomandibular field of view.
Good quality of the scan with no artefacts
CBCT scans with significant artifacts (e.g. metal artefact) affecting visualization of the mandible
Congenital craniofacial anomalies affecting the mandibular anatomy
Large pathologies that affect the mandible's dimensions and cause significant bone loss.
Trauma to the mandible
CBCT scans of patients, who have already been imaged for other treatment purposes, will be retrieved. Automatically segmented mandibles from these CBCT scans will be the population of the study.
Other
Retrospective
Not listed
Accuracy of AI in gender prediction
The accuracy of machine learning models (Artificial Neural Networks and Logistic Regression) in gender prediction. The predictions results are compared to the known gender from patient records which serve as the reference standard. The measuring unit is categorical as male or female. Accuracy, Precision, Recall (sensitivity), Specificity, F1-score, and a Confusion matrix will be used to evaluate the prediction models.
Accuracy of AI in anatomical point placement and linear measurements
Landmark placement and linear measurments of 3D UNet model versus radiologists' manual measurements will be compared. The measuring unit will be in millimeters (mm). Mean Euclidean Distance Error (in millimeters), and Root Mean Square Error (RMSE) will be used to assess the accuracy of the model in anatomical point placement in comparison to the radiologist's measurements
Click on the lead sponsor to view all of their trials.
