Prediction of hand‐wrist maturation stages based on cervical vertebrae images using artificial intelligence

Abstract

Objective: To predict the hand-wrist maturation stages based on the cervical vertebrae (CV) images, and to analyse the accuracy of the proposed algorithms. Settings and population:A total of 499 pairs of hand-wrist radiographs and lateral cephalograms of 455 orthodontic patients aged 6-18 years were used for developing the prediction model for hand-wrist skeletal maturation stages. Materials and Methods:The hand-wrist radiographs and the lateral cephalograms were collected from two university hospitals and a paediatric dental clinic. After identifying the 13 anatomic landmarks of the CV, the width-height ratio, widthperpendicular height ratio and concavity ratio of the CV were used as the morphometric features of the CV. Patients' chronological age and sex were also included as input data. The ground truth data were the Fishman SMI based on the hand-wrist radiographs. Three specialists determined the ground truth SMI. An ensemble machine learning methods were used to predict the Fishman SMI. Five-fold cross-validation was performed. The mean absolute error (MAE), round MAE and root mean square error (RMSE) values were used to assess the performance of the final ensemble model. Results:The final ensemble model consisted of eight machine learning models. The MAE, round MAE and RMSE were 0.90, 0.87 and 1.20, respectively. Conclusion:Prediction of hand-wrist SMI based on CV images is possible using machine learning methods. Chronological age and sex increased the prediction accuracy.An automated diagnosis of the skeletal maturation may aid as a decision-supporting tool for evaluating the optimal treatment timing for growing patients.

Inteligencia artificial en diagnóstico, pronóstico y planificación del tratamiento de alteraciones de la región cráneo-cérvico maxilofacial en ortodoncia. Revisión de la literatura

“…Los periodos de crecimiento y desarrollo mediante radiografías cárpales y las comparo con el análisis de una radiografía cefálica lateral por 24 modelos diferente de ANN; luego relacionaron el nivel de maduración y se pudo observar correlaciones significativamente positivas entre los datos de la radiografía carpal y los niveles de desarrollo del crecimiento vertebral cervical y la edad (p < 0,001) siendo el de mayor exactitud el modelo (17,19). Kim et al hiso la misma comparación usando 8 modelos de ML, "ayesianRidge", "LinearRegression, HuberRegressor", "SGD Regressor", "RandomForest Regressors", "TheilSen Regressor", "AdaBoost Regressor" y "LinearSVR" de los cuales las Medidas de error absoluto, Medidas de error absoluto redondeadas y raíz cuadrada de las medidas de error absoluto tuvieron 0.90, 0.87 y 1.20, respectivamente, dando un alto grado de correlación entre ambas, y con información adicional como la edad y el sexo se convierte en una herramienta muy útil en la toma de decisiones para la edad de tratamiento optima en pacientes en crecimiento (20).…”

Morphology-Aware Interactive Keypoint Estimation

“…Keypoint-based analysis of medical images is a widely used approach in clinical treatment [1,6,8,18,26]. Anatomical points obtained from an image can be utilized to measure significant features of body parts where keypoints are located.…”

“…Anatomical points obtained from an image can be utilized to measure significant features of body parts where keypoints are located. For example, vertex points of cervical vertebrae can be used to measure the concavity and height of the vertebrae, which correlates to skeletal growth [6,18]. Usually, doctors manually annotate such keypoints, but the process is costly, requiring The goal is to estimate (a) 13 keypoints on the cervical vertebrae, each of which determines vertebrae morphology.…”

“…Each image is annotated with 13 keypoints, which include the vertex points of the cervical vertebrae. The keypoints can be used to examine the skeletal growth of a patient by calculating anatomical measurements such as the concavity of the vertebrae[6,18]. The AASCE dataset[26] contains spinal anterior-posterior X-ray images (352/128/128 for training/validation/test).…”

Artificial intelligence and machine learning in orthodontics

“…Two groups of Kim et al and Kök et al evaluated the automated diagnosis of skeletal maturation using cervical vertebrae images on cephalometry and that AI‐based decision supporting is available for evaluating the optimal timing of treatment for growing orthodontic patients 18,19 . Both groups also found that that additional patient information increased the accuracy of the prediction.…”