Related Experiment Video
Updated: Sep 3, 2025
An Assessment Method and Toolkit to Evaluate Keyboard Design on Smartphones
Published on: October 5, 2020
Evaluation and comparison of smartphone application tracing, web based artificial intelligence tracing and
Delal Dara Kılınç1, Beyza Hancıoğlu Kırcelli2, Sanaz Sadry2
1Private Practice, Istanbul, Turkey.
Aim:
The aim of this study was to compare and evaluate the reliability of three different cephalometric assessment methods: Smartphone Application Tracing Method CephNinja (SATM), Web Based Artificial Intelligence (AI) Driven Tracing Method WebCeph (WATM) and Conventional Hand Tracing Method (CHTM).
Methods:
110 lateral cephalometric radiographs were enrolled in the study and 4 linear and 7 angular parameters were traced and measured by one examiner using CephNinja, WebCeph and conventional hand tracing methods. Independent-samples Kolmogorov-Smirnov and Shapiro Wilks tests were used to compare the mean values of intra-examiner differences. Both intra-method and inter-method correlations were evaluated.
Results:
There were statistically significant differences between the methods in terms of SNA (p:0.003; p < 0.05); SNB measurements (p:0.001; p < 0.05); SN-MP angle (p:0.001; p < 0.05); U1-SN angle (p:0.001; p < 0.05); L1-NB(mm) (p:0.007; p < 0.05) and E Line-Upper Lip(mm) measurements (p:0.013; p < 0.05). All intra-method correlation coefficients are 80% and above. In terms of inter-method coefficients the lowest coefficient of agreement is 0.170 and it is the coefficient of agreement between CHTM and SATM for measurement of U1-NA. The highest coefficient of agreement is 0.884 which is between WBTM and SATM for SNB measurement.
Conclusions:
The null hypothesis was rejected. There were statistically and clinically significant differences in various measurements among groups.
