Title

Impact of artificial intelligence (AI) on predicting marginal fit and aesthetic outcomes for custom implant abutments

Authors

Sarathchandra Govind Raj^1,*, Venkata Raghavan², Rahul Sharma³, Arunagiri Karunanithi⁴, Suma Janya⁵ & Minu Raju⁶

Affiliation

¹Department of Prosthodontics, Rajas Dental College and Hospital, Kavalkinaru Junction, Tirunelveli-627105, Tamil Nadu, India; ²Department of Periodontics, Chettinad Dental College & Research Institute, Kelambakkam, Kanchipuram -603103, Tamil Nadu, India; ³Department of Oral Medicine and Radiology, Maharana Pratap Dental College and Research Centre, Gwalior, Madhya Pradesh-474006, India; ⁴Department of Periodontics, Mahatma Gandhi Postgraduate Institute of Dental Science, Indira Nagar, Gorimedu, Puducherry, India; ⁵Department of Prosthodontics, Faculty of Dental Sciences, RUAS (Ramaiah University of Applied Sciences), MSRIT Educampus New BEL Road, Bengaluru-560064, India; ⁶Department of Prosthodontics, Mar Baselios Dental college, Kothamangalam, Ernakulam, Kerala-686691, India; *Corresponding author

Email

Sarath Chandra Govind Raj - E-mail: sarathgraj007@gmail.com; Phone: +91 9176121962
Venkata Raghavan - E-mail: dr.venkatraghavan94@gmail.com; Phone: +91 73061 34196
Rahul Sharma - E-mail: rahulsharma.vmdh@gmail.com; Phone: +91 94253 40325
Arunagiri Karunanithi - E-mail: giriarun133@gmail.com; Phone: +91 9486767181
Suma Janya - E-mail: sumajanya007@gmail.com; Phone: +91 97390 04015
Minu Raju - E-mail: minuraju31@gmail.com; Phone: +91 94969 52077

Article Type

Research Article

Date

Received October 1, 2025; Revised October 31, 2025; Accepted October 31, 2025, Published October 31, 2025

Abstract

The integration of artificial intelligence into implant prosthodontics enhances the precision of pre-fabrication predictions for clinical success. In this study, a custom convolutional neural network (CNN) model achieved a prediction accuracy of 93.5% for marginal fit within a 25 μm tolerance and 87.6% concordance with clinical evaluations of custom implant abutment aesthetics. The AI-estimated mean spatial gap (78.6±18.2 μm) closely approximated the actual measurement (82.3±21.5 μm), with strong correlations observed between predicted and actual outcomes for both fit (r = 0.89) and aesthetic appeal (r = 0.82–0.85). Thus, we show the potential of AI as a preventive quality assessment tool in implant prosthodontics, capable of minimizing adjustments and remakes while improving overall clinical success rates.

Keywords

Artificial intelligence; dental implants; marginal fit; custom abutments; aesthetic outcomes; deep learning

Citation

Raj et al. Bioinformation 21(10): 3962-3967 (2025)

Edited by

A Prashanth

ISSN

0973-2063

Publisher

Biomedical Informatics

License

This is an Open Access article which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. This is distributed under the terms of the Creative Commons Attribution License.