Academic Journal
AI-Driven Design of Auxiliary Devices to Improve Intraoral Scanning Accuracy in Complete-Arch Implant Cases.
| Τίτλος: | AI-Driven Design of Auxiliary Devices to Improve Intraoral Scanning Accuracy in Complete-Arch Implant Cases. |
|---|---|
| Συγγραφείς: | Pan Y, Chen W, Moy P, Pow EHN |
| Πηγή: | The International journal of oral & maxillofacial implants [Int J Oral Maxillofac Implants] 2026 Sep 30; Vol. 41 (5), pp. 635-642. |
| Τύπος έκδοσης: | Journal Article |
| Γλώσσα: | English |
| Στοιχεία περιοδικού: | Publisher: Quintessence Pub. Co Country of Publication: United States NLM ID: 8611905 Publication Model: Print Cited Medium: Internet ISSN: 1942-4434 (Electronic) Linking ISSN: 08822786 NLM ISO Abbreviation: Int J Oral Maxillofac Implants Subsets: MEDLINE |
| Imprint Name(s): | Original Publication: [Lombard, Ill. : Quintessence Pub. Co., c1986- |
| Ιατρικοί όροι (MeSH): | Dental Arch*/diagnostic imaging , Dental Arch*/surgery , Imaging, Three-Dimensional* , Dental Implants* , Deep Learning* , Artificial Intelligence* , Dental Prosthesis Design*, Humans ; Neural Networks, Computer ; Computer-Aided Design ; Multilayer Perceptrons |
| Περίληψη: | Purpose: To develop auxiliary devices for intraoral scanning (IOS) of complete-arch implant cases using a deep-learning artificial intelligence (AI) model. Materials and Methods: A total of 338 3D-imaging data sets were collected from a dental laboratory. Of these, 300 sets of complete dental arches were used for training, 28 sets were used for validation, and 10 edentulous arches with four to six dental implants were used for testing. Auxiliary devices with landmarks placed in between implants to aid in image stitching were manually designed and used as a control. A multilayer perceptron artificial neural network was employed to predict the positions of the landmarks, with normalized implant coordinates being used for input and landmark coordinates being used for output. The model was validated and evaluated using the test set to assess the fit of the base as well as the surface area of the landmarks. Results: The bounding box loss for the training and validation sets converged to 0.02 and 0.01, respectively, which indicated high precision in landmark position prediction. The objectness loss stabilized at 0.05 for the training set and 0.03 for the validation set, confirming the model's robust detection capability. The root mean square (RMS) of the device base was 0.117 ± 0.053 mm, significantly smaller than the clinical threshold of 0.300 mm (P < .001). The surface area of the AIgenerated device landmarks (762.0 ± 141.7 mm2) was significantly smaller than that of the manually designed control (1,307.1 ± 286.1 mm2, P = .001). Conclusions: The AI model demonstrated exceptional performance in this study. The base of the AI-generated auxiliary device fit well with the edentulous region, though its landmark teeth were smaller than those in the manually designed control. |
| Contributed Indexing: | Keywords: deep learning; dental implants; edentulism; intraoral scanning |
| Substance Nomenclature: | 0 (Dental Implants) |
| Entry Date(s): | Date Created: 20250716 Date Completed: 20260930 Latest Revision: 20260930 |
| Update Code: | 20261001 |
| DOI: | 10.11607/jomi.11415 |
| PMID: | 40668357 |
| Βάση Δεδομένων: | MEDLINE |
| ISSN: | 1942-4434 |
|---|---|
| DOI: | 10.11607/jomi.11415 |