Three-Dimensional Postoperative Results Prediction for Orthognathic Surgery through Deep Learning-Based Alignment Network

Deep learning-based prediction of osseointegration for dental implant using plain radiography

BACKGROUND

In this study, we investigated whether deep learning-based prediction of osseointegration of dental implants using plain radiography is possible.

METHODS

Panoramic and periapical radiographs of 580 patients (1,206 dental implants) were used to train and test a deep learning model. Group 1 (338 patients, 591 dental implants) included implants that were radiographed immediately after implant placement, that is, when osseointegration had not yet occurred. Group 2 (242 patients, 615 dental implants) included implants radiographed after confirming successful osseointegration. A dataset was extracted using random sampling and was composed of training, validation, and test sets. For osseointegration prediction, we employed seven different deep learning models. Each deep-learning model was built by performing the experiment 10 times. For each experiment, the dataset was randomly separated in a 60:20:20 ratio. For model evaluation, the specificity, sensitivity, accuracy, and AUROC (Area under the receiver operating characteristic curve) of the models was calculated.

RESULTS

The mean specificity, sensitivity, and accuracy of the deep learning models were 0.780-0.857, 0.811-0.833, and 0.799-0.836, respectively. Furthermore, the mean AUROC values ranged from to 0.890-0.922. The best model yields an accuracy of 0.896, and the worst model yields an accuracy of 0.702.

CONCLUSION

This study found that osseointegration of dental implants can be predicted to some extent through deep learning using plain radiography. This is expected to complement the evaluation methods of dental implant osseointegration that are currently widely used.

Three-Dimensional Comparison of the Maxillary Surfaces through ICP-Type Algorithm: Accuracy Evaluation of CAD/CAM Technologies in Orthognathic Surgery

PURPOSE

This retrospective study aims to compare the accuracy of two different CAD/CAM systems in orthognathic surgery. The novelty of this work lies in the method of evaluating the accuracy, i.e., using an Iterative Closest Point (ICP) algorithm, which matches a pair of 2D or 3D point clouds with unknown dependencies of the transition from scan s_(k) to scan s_(k+1).

METHODS

The study population was composed of ten patients who presented to the Maxillofacial Surgery Department of the University "Sapienza" of Rome for the evaluation and management of skeletal malocclusions. The patients were divided into two groups, depending on the technique used: group 1: splintless group (custom-made cutting guide and plates); group 2: splint group (using a 3D-printed splint). STL files were imported into Geomagic^® Control X™ software, which allows for comparison and analysis using an ICP algorithm. The RMSE parameter (3D error) was used to calculate the accuracy. In addition, data were compared in two different patient subgroups. The first subgroup only underwent a monobloc Le Fort I osteotomy (p-value = 0.02), and the second subgroup underwent a Le Fort I osteotomy associated with a segmental osteotomy of the maxilla (p-value = 0.23).

RESULTS

Group 1 showed a 3D error of 1.22 mm ± SD 0.456, while group 2 showed a 3D error of 1.63 mm ± SD 0.303. These results have allowed us to compare the accuracy of the two CAD/CAM systems (p-value = 0.09).

CONCLUSIONS

The ICP algorithm provided a reproducible method of comparison. The splintless method would seem more accurate (p-value = 0.02) in transferring the surgical programming into the operating room when only a Le Fort I osteotomy is to be performed.