← Back to Publications List

Domain-Specific NLP for Personalized Radiation Treatment Pathways with LLM Fine-Tuning

Students & Supervisors

Student Authors
Apurbo Biswas
Bachelor of Science in Computer Science & Engineering, FST
Md. Owafeeuzzaman Patwary
Bachelor of Science in Computer Science & Engineering, FST
Morzia Momo
Bachelor of Science in Computer Science & Engineering, FST
Sumaiya Tasnim
Bachelor of Science in Computer Science & Engineering, FST
Supervisors
Tofayet Sultan
Lecturer, Faculty, FST
Abhijit Bhowmik
Associate Professor, Faculty, FST

Abstract

General-purpose large language models (LLMs) such as GPT-4 and LLaMA-2 exhibit strong general reasoning capabilities but struggle to interpret domain-specific biomedical terminology and complex oncological semantics, often producing factual inaccuracies or hallucinations. To address these limi- tations, this study introduces the Radiation Oncology Model (ROM) a domain-adapted NLP system for personalized radiation treatment planning. ROM is built upon BioGPT and fine- tuned using the Radiation Oncology NLP Database (ROND) through a Retrieval-Augmented Fine-Tuning (RAG-FT) approach employing Low-Rank Adaptation (LoRA). The model integrates multimodal patient data, including clinical narratives, genomic information, and imaging features, to generate individualized radiotherapy pathways. Experimental results demonstrate signif- icant performance improvements, achieving an average F1-score of 84.09%, surpassing baseline models such as RadOnc-GPT (≈79.06%). ROM also exhibits reduced hallucination rates and improved contextual grounding aligned with clinical guidelines. Qualitative analyses confirm enhanced interpretability, factual consistency, and entity recognition across biomedical categories. These findings highlight that domain-specific fine-tuning and retrieval grounding substantially enhance clinical reliability and reasoning in AI-driven oncology. The proposed ROM framework establishes a pathway toward clinically interpretable and trust- worthy LLMs for precision radiotherapy and broader personal- ized oncology applications.

Keywords

Radiation oncology natural language process- ing large language models retrieval-augmented fine-tuning BioGPT precision oncology.

Publication Details

  • Type of Publication:
  • Conference Name: 2025 IEEE International Women in Engineering (WIE) Conference on Electrical and Computer Engineering 2025
  • Date of Conference: 21/12/2025 - 21/12/2025
  • Venue: Long Beach Hotel, Cox's Bazar, Bangladesh
  • Organizer: Dr. M. Imamul Hassan Bhuiyan, Professor, Department of EEE, BUET, Bangladesh
View on Publisher