Biomedical Image Processing for Age Measurements of Intact Teeth

Abstract

With increasing age the roots of teeth undergo sclerosis. The degree of dental root sclerosis can be demonstrated visually if light is transmitted through the specimen. However, this resultant image is only a two-dimensional (2-D) visualization which misrepresents what in truth is a three-dimensional (3-D) characteristic. We have described an image acquisition and computer processing system for imaging intact teeth, with special reference to the root transparency, which tends to progress from the root apex towards the neck of the teeth as a function of age. The components of our system involve: (1) an energy source with light transmission through table-mounted dental specimens; (2) an image processor with a digitizer; (3) a step motor with a holder to rotate the tooth specimen; and (4) a software package to computerize and reconstruct the sectional digital images. After rotating the position of the specimen, while at each rotation obtaining a 2-D image of the sample, we then can reconstruct the true 3-D cross-sectional or longitudinal morphology or both from these 2-D images. With this new approach, the reconstructed dental images represent segments from different angles of the tooth specimen. Picture element values in each image, quantitatively indicate the optical density, expressing the age dependent pattern of the 3-D anatomy in toto.

Age estimation using microfocus X-ray computed tomography of lower premolars

“…Table 2 shows Peason's correlation coefficients between age and variables (PTVR [1][2][3][4]. The ratio at the region of apical one third of the root (L 4 ) correlated least whereas the coronal one third of the root (L 2 ) correlated best with age both in lower first and second premolars.…”

“…The regression model with the highest value of R* 2 must be the best after decreasing the number of the independent variables by stepwise procedure. The regression model (1) for lower first premolars and model (2) for lower second premolars were obtained as the final best equations for predicting age. By comparing R* 2 values, the accuracy of estimating age obtained using model (2) for lower second premolars (R* 2 = 0.685) was slightly better than model (1) for lower first premolars (R* 2 = 0.617).…”

“…The regression model (1) for lower first premolars and model (2) for lower second premolars were obtained as the final best equations for predicting age. By comparing R* 2 values, the accuracy of estimating age obtained using model (2) for lower second premolars (R* 2 = 0.685) was slightly better than model (1) for lower first premolars (R* 2 = 0.617). This was also reinforced by the results for the value of mean predicting error for lower first premolars which was higher than that for lower second premolars.…”

“…Subsequently, there have been many articles detailing age estimation methods based on morphological measurements of changes as an indicator of age [2][3][4][5][6][7][8][9][10][11][12][13][14][15][16]. In these articles secondary dentine deposits have been regarded as a valuable age factor and have been studied by several methods requiring preparation of microscopic sections or radiographic techniques.…”

“…Using the reconstructed images, the tooth and pulp volumes at four levels (crown area (L 1 ); coronal one third (L 2 ); midroot (L 3 ), and apical one third of the root (L 4 )) were measured. Mean values of the pulp-tooth volume ratio (PTVR [1][2][3][4] ) were compared between the six age groups and these values were used as the variables for the multiple regression analysis.…”