HOME   |    PDF   |   


Title

AI-based prediction of implant prosthesis micromotion under simulated mastication: An in vitro study

 

Authors

Nirma Yadav 1, Anand Kumar 2, Inderpreet Kaur 3, Vijay Parmar 4,*, Shubham Bhuktar 5 & Gaurav Chandra 6

 

Affiliation

1Department of Dentistry, Maa Vindhyavasini Autonomous State Medical College, Mirzapur, Uttar Pradesh, India; 2Department of Dentistry, Moti Lal Nehru Medical College, Prayagraj, Uttar Pradesh, India; 3Department of Prosthodontics, Adesh Institute of Dental Sciences and Research, Punjab, India; 4Department of Dentistry, AIIMS Rajkot, Gujarat, India; 5Department of Prosthodontics, Crown and Bridge and Implantology, Dr Rajesh Ramdasji Kambe Dental College and Hospital, Akola, Maharashtra, India; 6Department of Prosthodontics and Crown and Bridge, Chandra Dental College Safedabad, Barabanki, Uttar Pradesh, India; *Corresponding author

 

Email

Nirma Yadav - E-mail: nirmay90@gmail.com
Anand Kumar - E-mail: anandkmr901@gmail.com
Inderpreet Kaur - E-mail: inderbathinda876@gmail.com
Vijay Parmar - E-mail: vijayparmar8597@gmail.com
Shubham Bhuktar - E-mail: shubh.bhuktar@gmail.com
Gaurav Chandra - E-mail: grv.chandra@gmail.com
 

Article Type

Research Article

 

Date

Received October 1, 2026; Revised October 31, 2026; Accepted October 31, 2026, Published October 31, 2026

 

Abstract

Cyclic loading can alter implant-abutment settling, screw preload and prosthesis micromotion, while instability is governed by interacting mechanical variables. Therefore, it is of interest to develop and test a machine-learning model for predicting prosthesis instability after simulated masticatory cycling in 120 implant-supported zirconia crowns with internal conical, internal hex, or external hex connections. Digital image correlation, reverse-torque assessment, marginal-gap measurements and 200,000 loading cycles at 100 N were used and gradient boosting incorporated connection design, insertion torque, preload loss, marginal gap, baseline micromotion and cycle-stage features. Thirty assemblies met the instability definition; the test-set model achieved an AUC of 0.93, sensitivity of 86.7%, specificity of 88.9% and Brier score of 0.10, with baseline micromotion, post-settling torque loss and connection type as the strongest predictors. Thus, machine learning identified fatigue-related instability in this controlled model, but independent external validation is required before mechanical decision support is considered.

 

Keywords

Artificial intelligence (AI), implant prosthesis, micromotion, cyclic loading, screw loosening, machine learning (ML)

 

Citation

Yadav et al. Bioinformation 22(10): 7547-7551 (2026)

 

Edited by

Ashwini Dhopte

 

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.