Evaluation of soft tissue profile with artificial intelligence algorithms developed with deep learning method on profile images

Aim: Today, new artificial intelligence technologies have been developed based on the principles of imitating the functioning of the human brain. The aim of this study is to investigate the success of anatomical point detection using a special artificial intelligence algorithm on profile images. Material and Method: The data set of our study is to create the anatomical points marked using a special artificial intelligence algorithm on the records obtained from the profile images of 1000 patients taken before orthodontic treatment. Labeling of spots on images was done using CranioCatch labeling software. The training of the artificial intelligence model was carried out using 300 Epochs with the CNN-based deep learning method applied to PyTorch. Results: In the artificial intelligence model trained in the study, the highest SDR values were found at soft tissue A and soft tissue B points. The lowest SDR value was observed at the Glabella point. The SDR rate showed values above 90%, except for 6 points in a total of 23 points marked at a 4 mm interval. Conclusion: Our study is very important for the development of deep learning-based facial soft tissue analysis systems in the future. It is thought that systems created using artificial intelligence algorithms will serve as a decision support mechanism that will save physicians time in their clinical routine.At the same time, it is estimated that during the determination of anatomical points, it will be very helpful in minimizing the differences between observers and inconsistencies that may occur in the evaluations of the observers at different times. Keywords: Anatomic Landmarks, Artificial intelligence, Deep learning, Profile image