# Ontology-Based Concept Distillation for Radiology Report Retrieval and Labeling

## TL;DR

This paper introduces an ontology-based method for radiology report retrieval that leverages UMLS concepts for interpretable and semantically meaningful similarity measures, outperforming traditional embedding-based approaches in medical imaging tasks.

## Contribution

It presents a novel ontology-driven approach for report comparison using UMLS concepts, enhancing interpretability and performance in radiology report retrieval and labeling.

## Key findings

- Outperforms state-of-the-art embedding methods in radiograph classification.
- Improves retrieval accuracy in long-tail medical imaging tasks.
- Provides ontology-backed disease labels for downstream tasks.

## Abstract

Retrieval-augmented learning based on radiology reports has emerged as a promising direction to improve performance on long-tail medical imaging tasks, such as rare disease detection in chest X-rays. Most existing methods rely on comparing high-dimensional text embeddings from models like CLIP or CXR-BERT, which are often difficult to interpret, computationally expensive, and not well-aligned with the structured nature of medical knowledge. We propose a novel, ontology-driven alternative for comparing radiology report texts based on clinically grounded concepts from the Unified Medical Language System (UMLS). Our method extracts standardised medical entities from free-text reports using an enhanced pipeline built on RadGraph-XL and SapBERT. These entities are linked to UMLS concepts (CUIs), enabling a transparent, interpretable set-based representation of each report. We then define a task-adaptive similarity measure based on a modified and weighted version of the Tversky Index that accounts for synonymy, negation, and hierarchical relationships between medical entities. This allows efficient and semantically meaningful similarity comparisons between reports. We demonstrate that our approach outperforms state-of-the-art embedding-based retrieval methods in a radiograph classification task on MIMIC-CXR, particularly in long-tail settings. Additionally, we use our pipeline to generate ontology-backed disease labels for MIMIC-CXR, offering a valuable new resource for downstream learning tasks. Our work provides more explainable, reliable, and task-specific retrieval strategies in clinical AI systems, especially when interpretability and domain knowledge integration are essential. Our code is available at https://github.com/Felix-012/ontology-concept-distillation