A GENERATIVE ARTIFICIAL INTELLIGENCE POWERED CLINICAL DECISION SUPPORT SYSTEM FOR PERSONALIZED TREATMENT PLANNING IN PROSTHODONTICS

Generative Artificial Intelligence (GenAI), Clinical Decision Support System (CDSS), Prosthodontics, Personalized Treatment Planning, Large Language Models (LLMs)

1*Pranavanand Satyamurthy,
Vallurupalli Nageshwara Rao Vignana Jyothi
Institute of Engineering and Technology, Hyderabad,
Telangana, India. pranavanand_s@vnrvjiet.in

2Shruthishree S. H.,
Associate Professor, Department of Computer
Science and Engineering (Artificial Intelligence
& Machine Learning), Global Academy of
Technology, Bengaluru, Karnataka, India.
sh.shruthishree@gat.ac.in

3T. Pradeep
Associate Professor, Department of Civil
Engineering, Kongu Engineering College,
Perundurai, Erode – 638060, Tamil Nadu,
India. pradeep@kongu.ac.in

4Venkata Vaisali Vavilala
Assistant Professor, Department of Computer
Science, Vignan Degree & P.G. College,
Guntur, Andhra Pradesh – 522009, India.
vaisalivavilala@gmail.com

5A. Swathi
Assistant Professor, Department of Statistics,
Vignan Degree & P.G. College, Guntur, Andhra
Pradesh – 522009, India.
akurathi.swathi@gmail.com

6Mutyala Suresh
Associate Professor, Koneru Lakshmaiah
Education Foundation (KL Deemed to be
University), Guntur District, Andhra Pradesh –
522302, India. msphd@kluniversity.in

Received:10-06-2026
Revised:17-07-2026
Accepted:28-07-2026

European Journal of Prosthodontics and Restorative Dentistry (2026) 34 (7s), 538–553

A GENERATIVE ARTIFICIAL INTELLIGENCE POWERED CLINICAL DECISION SUPPORT SYSTEM FOR PERSONALIZED TREATMENT PLANNING IN PROSTHODONTICS

Generative Artificial Intelligence (GenAI) is revolutionizing the field of prosthodontics, where GenAI is used for providing intelligent decision support for treatment planning and prosthesis choice with respect to specific patients. In this retrospective multicenter validation study, the proposed Evidence grounded hybrid clinical reasoning framework (EHCRF) combines multimodal patient information, including demographics, medical history, intraoral examination, radiographs, occlusion, digital impressions, and preferences of clinicians in order to recommend personalized treatment plans. Specifically, in the proposed framework, a multimodal explainable AI system is developed, which leverages transformer-based large language models (LLMs), retrieval-augmented generation (RAG) based on semantic vectors, explainable artificial intelligence (XAI), multimodal data fusion, and prosthodontic evidence-based guidelines in order to provide transparent and context-sensitive decisions for clinical decision making. In terms of validation of the proposed framework, 1,248 cases were assessed in the validation set by 25 experienced prosthodontists compared to CAD/CAM-based treatment planning driven by experts. Consequently, the proposed approach resulted in 96.8% diagnostic accuracy (95% confidence interval [CI]: 95.7-97.9%), 95.9% treatment recommendation agreement, 2.3 seconds average response time, and 4.8/5 clinicians' satisfaction (5-point Likert scale). All the above results showed statistically significant performance gain (p < 0.001). The main contribution is the creation of the evidence-based hybrid model of clinical reasoning that combines multimodal patient data, augmented reasoning with retrieval methods, explainable AI, and large language models into one prosthodontic decision support system

•••••••••••••••••••••••••••••••• ejprd.org - Published by Riset Publication Services LLC

EJPRD

Copyright ©2026 by Riset Publication Services LLC