| Autor: |
Sung-Hoon Han, Jisup Lim, Jun-Sik Kim, Jin-Hyoung Cho, Mihee Hong, Minji Kim, Su-Jung Kim, Yoon-Ji Kim, Young Ho Kim, Sung-Hoon Lim, Sang Jin Sung, Kyung-Hwa Kang, Seung-Hak Baek, Sung-Kwon Choi, Namkug Kim |
| Předmět: |
|
| Zdroj: |
Korean Journal of Orthodontics; Jan2024, Vol. 54 Issue 1, p48-58, 11p |
| Abstrakt: |
This article discusses a study that evaluated the accuracy of a cascaded convolutional neural network (CNN) algorithm in identifying landmarks and measuring deviations in posteroanterior (PA) cephalograms. The study found that the algorithm had a clinically acceptable level of error and high accuracy in landmark identification. The document also discusses the training and validation process of the algorithm, as well as the measurement errors and statistical analysis performed. Another document discusses the accuracy of artificial intelligence (AI) in identifying landmarks on PA cephalometric images. The AI demonstrated high accuracy for certain landmarks but lower accuracy for others, and performed better than first-year orthodontic residents. The text also discusses the use of AI algorithms for the identification and measurement of cephalometric landmarks in orthodontic research, highlighting the need for further studies to compare the accuracy of examiners with different levels of clinical experience. [Extracted from the article] |
| Databáze: |
Supplemental Index |
| Externí odkaz: |
|
