Search papers, labs, and topics across Lattice.
The paper introduces BTReport, a framework for brain tumor radiology report generation (RRG) that uses deterministic feature extraction from imaging data to generate natural language reports. This approach separates image analysis from report composition, using large language models (LLMs) solely for syntactic structuring and narrative formatting, thereby enhancing interpretability and reducing hallucinations. Experiments demonstrate that the extracted features are predictive of clinical outcomes like survival and IDH mutation status, and the generated reports exhibit closer alignment with reference clinical reports compared to existing RRG baselines.
Ditch the end-to-end vision-language model for radiology reports: BTReport shows that deterministic feature extraction paired with LLM-based formatting yields more clinically relevant and interpretable reports.
Recent advances in radiology report generation (RRG) have been driven by large paired image-text datasets; however, progress in neuro-oncology has been limited due to a lack of open paired image-report datasets. Here, we introduce BTReport, an open-source framework for brain tumor RRG that constructs natural language radiology reports using deterministically extracted imaging features. Unlike existing approaches that rely on large general-purpose or fine-tuned vision-language models for both image interpretation and report composition, BTReport performs deterministic feature extraction for image analysis and uses large language models only for syntactic structuring and narrative formatting. By separating RRG into a deterministic feature extraction step and a report generation step, the generated reports are completely interpretable and less prone to hallucinations. We show that the features used for report generation are predictive of key clinical outcomes, including survival and IDH mutation status, and reports generated by BTReport are more closely aligned with reference clinical reports than existing baselines for RRG. Finally, we introduce BTReport-BraTS, a companion dataset that augments BraTS imaging with synthetically generated radiology reports produced with BTReport. Code for this project can be found at https://github.com/KurtLabUW/BTReport.
