Back

Leveraging Large Language Models to Direct Automated PET/CT Tumour Segmentation in Retrospective Data: an Agentic Framework Method

Bonney, L. M.; Brown, S. L.; Bradley, K. M.; Hassan, A. B.; McGowan, D. R.

2025-11-30 radiology and imaging
10.1101/2025.11.27.25339233 medRxiv
Show abstract

Automated medical image segmentation using deep learning requires large labelled datasets, presenting barriers for rare cancers like sarcoma. We developed an agentic framework integrating LLM analysis of radiologist reports with nnUNet segmentation for 18F-FDG PET/CT imaging data (N=60, 134 studies), aiming to improve automated tumour segmentation for retrospective data. LLM interpretation was optimised in an expanded report dataset (N=91, 226 studies). The framework aimed to first screen for disease presence, then tested localising tumours to body regions mapped via automated organ segmentation, enabling targeted image cropping before segmentation. A baseline nnUNet was trained for comparison and achieved a mean Dice of 0.49. The disease classification framework attained mean Dice 0.62, correctly identifying 9/11 disease-free cases versus 6/11 for baseline. This approach demonstrates potential for leveraging routinely collected clinical text data to enhance medical imaging research in rare diseases.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.