Automated Generation of Dental Treatment Plans with LLM and RAG

Overview

The project involved designing and implementing an AI system for automated generation of dental treatment plans by combining classical machine learning, Retrieval-Augmented Generation (RAG), and large language models (LLM). The solution was developed around medical datasets, doctor-validated clinical rules, and human-in-the-loop feedback to improve quality, safety, and practical adoption in clinical workflows.

Achievements

Designed a robust medical AI architecture that combined LLMs, RAG, and structured clinical knowledge for treatment planning support. Built domain-specific data pipelines and introduced continual learning from clinician feedback. Strengthened medical relevance and safety by incorporating doctor-validated rules and constraints into the workflow.

Responsibilities

• Designed and implemented a hybrid AI system combining classical ML, RAG, and LLM components.
• Built ETL pipelines for medical data cleaning, normalization, and augmentation.
• Developed a medical knowledge base with clinician-validated rules and constraints.
• Created prompts and system logic for domain-specific LLM use cases.
• Implemented LoRA fine-tuning on specialized medical datasets.
• Built a continual learning pipeline driven by doctor corrections and human-in-the-loop review.
• Collaborated with doctors and stakeholders to align the system with clinical needs.

Technologies Used