Evaluation of Reliability and Validity of Linear and Angular Parameters via Analog and Digital Methods: A Lateral Cephalometric Analysis

The study aimed at determining the variability of landmark identifications and its reproducibility by manual and digital methods of cephalometric tracing.

Pre-treatment cephalograms of sixty subjects recommended for orthodontic treatment were taken in the natural head position with the teeth in maximum intercuspation and were traced by a single operator. Statistical analysis was performed, and the mean, standard deviations, and coefficients of variation were calculated for each parameter and imaging modality. For all the variables, the Shapiro-Wilk’s test for normality was conducted to identify the distribution of each variable. The Wilcoxon signed rank test was used where distribution was non-normal for at least one of the comparing variables and the repeated sample t-test was used where normal distribution was detected for both the variables.

For digital tracing, the coefficients of variation were less than 1, suggesting high reliability. For most manually traced images, the landmarks with high coefficients of variation were Porion, Basion, Sella, Point B, and Gonion. Coefficients of variation were invariably higher for the manual method than the digital method, implying readings were more consistent in the digital method for the above landmarks under observation. Out of the 29 parameters, the mean difference of sixteen of them was calculated to be statistically significant with each other, which shows that the digital method of cephalometric tracing can be relied upon for the reliability.

The digital method of landmark identification gives a more consistent reading.