• Journal of the Korean Society of Radiology
• /
• v.18 no.3
• /
• pp.267-273
• /
• 2024

Using a cloud-based vertex AI platform that can develop an artificial intelligence learning model without coding, this study easily developed an artificial intelligence learning model by the non-professional general public and confirmed its clinical applicability. Nine dental diseases and 2,999 root disease X-ray images released on the Kaggle site were used for the learning data, and learning, verification, and test data images were randomly classified. Image classification and multi-label learning were performed through hyper-parameter tuning work using a learning pipeline in vertex AI's basic learning model workflow. As a result of performing AutoML(Automated Machine Learning), AUC(Area Under Curve) was found to be 0.967, precision was 95.6%, and reproduction rate was 95.2%. It was confirmed that the learned artificial intelligence model was sufficient for clinical diagnosis.

• Journal of the Korean Society of Radiology
• /
• v.17 no.4
• /
• pp.515-521
• /
• 2023

The radiation dose received by the patient varies according to the tube current and time used during dental intraoral imaging. A large amount of tube current is required for image quality, but the radiation dose to the patient increases accordingly. Therefore, in this study, the optimal amount of tube current that can reduce the radiation dose received by the patient while securing the image quality was calculated through the evaluation of the image quality according to the tube current used during intraoral imaging through simulation. The average tube current, time, and tube voltage presented in the Guidelines for Diagnostic Reference Level for intraoral radiography were used as basic imaging conditions, and images were obtained when only the tube current was changed, and then the optimal tube current was compared and analyzed with the basic image quantity was calculated. Images were obtained by changing the tube current to 0.1, 0.5, 1, 2, 3, 4 and 5 mA under the basic conditions of 63 kV, 6 mA, and 0.29 s. The obtained image was evaluated for structural similarity index with the image taken under the condition of 6 mA using the ICY program. As a result, even under the condition of 0.5 mA tube current, the index of structural similarity with the image of 6 mA was evaluated to be high. Based on these results, it is considered that the radiation dose given to the patient can be greatly reduced if imaging is performed at 0.5 mA instead of 6 mA during dental intraoral imaging.