Leveraging Fine-Tuned Large Language Models for Interpretable Pancreatic Cystic Lesion Feature Extraction and Risk Categorization

July 26, 2025 Β· Declared Dead Β· πŸ› arXiv.org

πŸ‘» CAUSE OF DEATH: Ghosted
No code link whatsoever

"No code URL or promise found in abstract"

Evidence collected by the PWNC Scanner

Authors Ebrahim Rasromani, Stella K. Kang, Yanqi Xu, Beisong Liu, Garvit Luhadia, Wan Fung Chui, Felicia L. Pasadyn, Yu Chih Hung, Julie Y. An, Edwin Mathieu, Zehui Gu, Carlos Fernandez-Granda, Ammar A. Javed, Greg D. Sacks, Tamas Gonda, Chenchan Huang, Yiqiu Shen arXiv ID 2507.19973 Category cs.AI: Artificial Intelligence Cross-listed cs.CL, cs.IR Citations 1 Venue arXiv.org Last Checked 4 months ago
Abstract
Background: Manual extraction of pancreatic cystic lesion (PCL) features from radiology reports is labor-intensive, limiting large-scale studies needed to advance PCL research. Purpose: To develop and evaluate large language models (LLMs) that automatically extract PCL features from MRI/CT reports and assign risk categories based on guidelines. Materials and Methods: We curated a training dataset of 6,000 abdominal MRI/CT reports (2005-2024) from 5,134 patients that described PCLs. Labels were generated by GPT-4o using chain-of-thought (CoT) prompting to extract PCL and main pancreatic duct features. Two open-source LLMs were fine-tuned using QLoRA on GPT-4o-generated CoT data. Features were mapped to risk categories per institutional guideline based on the 2017 ACR White Paper. Evaluation was performed on 285 held-out human-annotated reports. Model outputs for 100 cases were independently reviewed by three radiologists. Feature extraction was evaluated using exact match accuracy, risk categorization with macro-averaged F1 score, and radiologist-model agreement with Fleiss' Kappa. Results: CoT fine-tuning improved feature extraction accuracy for LLaMA (80% to 97%) and DeepSeek (79% to 98%), matching GPT-4o (97%). Risk categorization F1 scores also improved (LLaMA: 0.95; DeepSeek: 0.94), closely matching GPT-4o (0.97), with no statistically significant differences. Radiologist inter-reader agreement was high (Fleiss' Kappa = 0.888) and showed no statistically significant difference with the addition of DeepSeek-FT-CoT (Fleiss' Kappa = 0.893) or GPT-CoT (Fleiss' Kappa = 0.897), indicating that both models achieved agreement levels on par with radiologists. Conclusion: Fine-tuned open-source LLMs with CoT supervision enable accurate, interpretable, and efficient phenotyping for large-scale PCL research, achieving performance comparable to GPT-4o.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

πŸ“œ Similar Papers

In the same crypt β€” Artificial Intelligence

Died the same way β€” πŸ‘» Ghosted