To investigate the association between the angles describing the emergence profile of an implant crown and metric data used to validate the digital design of a crown generated via three-dimensional generative adversarial neural network (3D GAN). A 3D GAN developed and trained by G. Kostadinov and A. Naydenov was used to generate 20 screw retained implant crowns replacing a missing 36 tooth. Their validation metric indicators, intersection over union (IoU), precision, recall, and F1 score, were recorded, totaling 80 metrics. The initial voxelized files were processed with “MeshMixer” software, followed by the creation of mediodistal and vestibulolingual longitudinal cuts using “ExoCad” software. For the analysis, “CorelDRAW” software was used, where the three angles, mucosal emergence angle (MEA), deep emergence angle (DA), and total contour angle (CA), were measured for each of the four examined surfaces (vestibular, lingual, mesial, and distal) per crown, totaling 240 examined angles. The overall median scores were 27.79° (MEA), 56.70° (DA), and 43.54 (CA). The overall median IoU, precision, recall, and F1 score was 0.835, 0.946, 0.942. and 0.91 across the 20 teeth. The average deviation from optimal values (equal to 1 for all visual metrics) is 0.16 for IoU, 0.074 for precision, 0.0875 for recall, and 0.09 for F1 score. Validation of AI must evolve to include clinical parameters beyond digital metrics. Visual parameters used to validate the design of implant crowns are not universal but rather supplementary measures for assessing the real clinical value of the newly generated crowns. Even in cases where IoU, precision, recall, and F1 score are not perfect (<1), clinically significant parameters like MEA still have an excellent mean score of 27.79, which is a predictive factor for peri-implant tissue health.
Naydenov et al. (Wed,) studied this question.