Development and expert radiologist validation of a custom pipeline for simplification of oncology radiology reports using large language model

1

A bilingual LLM-driven tool was developed to convert oncology radiology reports into simplified English and Hindi for better patient comprehension.
2

The tool was created using 100 CT reports from colorectal cancer patients and validated for accuracy and clarity by expert radiologists.
3

Average scores for accuracy, language clarity, and readability were high, indicating effective communication without compromising diagnostic integrity.
4

The Flesch Reading Ease score improved significantly from 49.2 to 73, demonstrating enhanced readability of the reports.
5

The study highlights the potential of AI in facilitating patient-centered communication in oncology while addressing emotional sensitivity.

Frontiers In Oncology

by Prerna Garg, Neil Agarwal, Abhisht Agarwal, Vasantha Kumar Venugopal, Mahendra KM, Bharat Gupta, Jain Manish, Jitin Goyal, Radhika Daga, Sunil Kumar Puri
June 23, 2026

1