Age estimation based on 3D pulp segmentation of first molars from CBCT images using U-Net

Abstract

Objective: To train a U-Net model to segment the intact pulp cavity of first molars and establish a reliable mathematical model for age estimation.

Methods: We trained a U-Net model by 20 sets of cone-beam CT images and this model was able to segment the intact pulp cavity of first molars. Utilizing this model, 239 maxillary first molars and 234 mandibular first molars from 142 males and 135 females aged 15-69 years old were segmented and the intact pulp cavity volumes were calculated, followed by logarithmic regression analysis to establish the mathematical model with age as the dependent variable and pulp cavity volume as the independent variable. Another 256 first molars were collected to estimate ages with the established model. Mean absolute error and root mean square error between the actual and the estimated ages were used to assess the precision and accuracy of the model.

Results: The dice similarity coefficient of the U-Net model was 95.6%. The established age estimation model was [Formula: see text] (V is the intact pulp cavity volume of the first molars). The coefficient of determination (R²), mean absolute error and root mean square error were 0.662, 6.72 years, and 8.26 years, respectively.

Conclusion: The trained U-Net model can accurately segment pulp cavity of the first molars from three-dimensional cone-beam CT images. The segmented pulp cavity volumes could be used to estimate the human ages with reasonable precision and accuracy.

Keywords: Age estimation; CBCT; First molar; Medical image segmentation; U-Net.

Figure 1.

The work flow of U-net training and pulp cavity segmentation. (a) The original DICOM data sets was imported into Dragonfly. (b) The voxel color gradient of CBCT images were normalized into [0, 1]. (c) The region of interest was cropped into nearly a smallest region only containing the complete first molar. (d) Slices of every CBCT data set from the sagittal plane as input frames and manually annotate pulp and the rest of tooth tissue as two different classes in every frame. (e) The annotated images were performed by data augmentation and input into the U-Net for encoding (downsampling) and decoding (upsampling) to accomplish training. (f) The trained U-Net output segmented images. (g) An example of the segmentation result.

Figure 2.

A schematic overview of the U-Net network.

Figure 3.

Differences of pulp cavity volumes obtained from automatic and manual segmentation against mean pulp cavity volumes obtained from automatic and manual segmentation. A, automatic segmentation; M, manual segmentation; SD, standard deviation.

Figure 4.

The distribution of volumes in each age group of the modeling group and the validation group.

Figure 5.

The scatter diagram of logarithmic regression analysis for the modeling group shows the relationship between pulp volumes and ages.

Figure 6.

Plots of actual age versus estimated age for the validation group.