Publications by authors named "Youngsung Yu"
Article Synopsis
- The study aimed to find the ideal amount of learning data needed for AI to identify cephalometric landmarks accurately.
- Researchers used 2400 cephalograms, with 2200 images for training the AI and 200 for testing, analyzing different combinations of learning data sizes and detection targets.
- Results showed that accuracy improved with more learning data, needing at least 2300 data sets to match human examiners' precision, highlighting the need for extensive data in AI development.
Objectives: To compare detection patterns of 80 cephalometric landmarks identified by an automated identification system (AI) based on a recently proposed deep-learning method, the You-Only-Look-Once version 3 (YOLOv3), with those identified by human examiners.
Materials And Methods: The YOLOv3 algorithm was implemented with custom modifications and trained on 1028 cephalograms. A total of 80 landmarks comprising two vertical reference points and 46 hard tissue and 32 soft tissue landmarks were identified.
Objective: To compare the accuracy and computational efficiency of two of the latest deep-learning algorithms for automatic identification of cephalometric landmarks.
Materials And Methods: A total of 1028 cephalometric radiographic images were selected as learning data that trained You-Only-Look-Once version 3 (YOLOv3) and Single Shot Multibox Detector (SSD) methods. The number of target labeling was 80 landmarks.