The accuracy of algorithms used by artificial intelligence in cephalometric points detection: a systematic review

Our findings suggest that CNNs represent the most promising AI form for detecting cephalometric landmarks in 2D lateral cranial teleradiography, offering lower error rates and higher reproducibility compared to other AI types reviewed. However, due to significant heterogeneity in study designs, data...

Descripción completa

Detalles Bibliográficos
Autores: Ribas-Sabartés, Júlia, Sánchez Molins, Meritxell, d'Oliveira, Nuno Gustavo
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2024
País:España
Institución:Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
Repositorio:Recercat. Dipósit de la Recerca de Catalunya
OAI Identifier:oai:recercat.cat:2445/218630
Acceso en línea:https://hdl.handle.net/2445/218630
Access Level:acceso abierto
Palabra clave:Cefalometria
Intel·ligència artificial
Ortodòncia
Cephalometry
Artificial intelligence
Orthodontics
Descripción
Sumario:Our findings suggest that CNNs represent the most promising AI form for detecting cephalometric landmarks in 2D lateral cranial teleradiography, offering lower error rates and higher reproducibility compared to other AI types reviewed. However, due to significant heterogeneity in study designs, data collection, and performance metrics, a definitive quantitative comparison was not feasible. While AI demonstrates faster and more reproducible results than manual tracing, no algorithms currently match the precision of human professionals. Future research should aim to standardize evaluation criteria and datasets to enable a more robust comparison of AI methods.