Reliability of Artificial Intelligence for Treatment Decision for Adult Skeletal Open Bite Patients

This study evaluated a specially designed AI model developed by a programmer using x-ray readings and corresponding treatment decisions from 70% of the cases (either orthodontic treatment only or orthodontic treatment with surgery). For the evaluation, we will use the remaining 30% of cases. "Subsequently, to assess its performance, the model was tested on the remaining 30% of cases. The programmer provided only the X-ray readings as input. The AI model was then tasked with classifying.

In this study, all patients were treated completely with a well-finished result by expert orthodontists. This study evaluates whether an artificial intelligence (AI) model can enhance treatment decisions for adult skeletal open bite cases by predicting the optimal intervention, either orthognathic surgery or camouflage, using cephalometric readings as input data. First, a total of 53 cases were analyzed, which were divided into two groups: 70% were allocated to the machine learning group (MLG), while the remaining 30% constituted the test group (TG). Cephalometric analysis for all patients was performed using Dolphin Imaging 11.5 Premium software, along with determining the appropriate treatment decision, either camouflage or orthognathic surgery. The data obtained from MLG serves as training data for the AI model to classify cases based on their cephalometric data, whether for camouflage or orthognathic surgery. The input data consisted of cephalometric readings along with a decision. Second, after machine learning, validation takes place to examine the ability of the machine to make decisions through some cases from MLG. The third step will evaluate the machine's ability to accurately determine case decisions based on cephalometric readings. The results produced by the machine will be compared to the actual decisions made, as all these cases were treated under the supervision of orthodontic professors.