arXiv 2509.044712025-10-02cs.CLcs.AI

MOSAIC：一种多语言、无类别依赖且计算高效的放射报告分类方法

MOSAIC: A Multilingual, Taxonomy-Agnostic, and Computationally Efficient Approach for Radiological Report Classification

Department of Computer Science, University of Copenhagen（哥本哈根大学计算机科学系）；Neurobiology Research Unit, Copenhagen University Hospital（哥本哈根大学医院神经生物学研究单位）；Unumed Aps（Unumed公司）；Department of Diagnostic Radiology, Copenhagen University Hospital（哥本哈根大学医院诊断放射学系）；Department of Clinical Medicine, University of Copenhagen（哥本哈根大学临床医学系）；Cerebriu A/S（Cerebriu公司）；Institute for Human Genetics, Medical University of Innsbruck（因斯布鲁克医学大学人类遗传学研究所）

Alice Schiavone, Marco Fraccaro, Lea Marie Pehrson, Silvia Ingala, Rasmus Bonnevie, Michael Bachmann Nielsen, Vincent Beliveau, Melanie Ganz, Desmond Elliott

备注：: 8 pages, 14 pages including references and appendix. 9 figures. Preprint
发表信息：: Proceedings of the ClinicalNLP Workshop at LREC 2026

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

MOSAIC通过紧凑开放模型实现多语言、无类别依赖的放射报告分类，无需大量标注数据，且在多种影像模态和标签体系上表现优异，达到专家水平性能。

摘要：

放射学报告包含丰富的临床信息，可用于训练影像模型而无需依赖昂贵的手动标注。然而，现有方法面临关键限制：基于规则的方法难以处理语言多样性，监督模型需要大量标注数据集，而近期基于LLM的方法依赖封闭源或资源密集型模型，不适合临床使用。此外，当前解决方案大多局限于英语和单模态、单类别数据集。我们介绍了MOSAIC，一种多语言、无类别依赖且计算高效的放射报告分类方法。基于紧凑的开放访问语言模型（MedGemma-4B），MOSAIC支持零/少样本提示和轻量级微调，可在消费级GPU上部署。我们在英语、西班牙语、法语和丹麦语的七个数据集上评估MOSAIC，涵盖多种影像模态和标签体系。该模型在五个胸部X光数据集上达到平均宏F1分数88，接近或超过专家水平性能，同时仅需24GB GPU内存。通过数据增强，仅需80个标注样本即可在丹麦报告上达到加权F1分数82，相比完整1600样本训练集的86分。MOSAIC为临床环境中大型或专有LLM提供了实用替代方案。代码和模型是开源的。我们邀请社区在新语言、类别和模态上评估和扩展MOSAIC。

英文摘要：

Radiology reports contain rich clinical information that can be used to train imaging models without relying on costly manual annotation. However, existing approaches face critical limitations: rule-based methods struggle with linguistic variability, supervised models require large annotated datasets, and recent LLM-based systems depend on closed-source or resource-intensive models that are unsuitable for clinical use. Moreover, current solutions are largely restricted to English and single-modality, single-taxonomy datasets. We introduce MOSAIC, a multilingual, taxonomy-agnostic, and computationally efficient approach for radiological report classification. Built on a compact open-access language model (MedGemma-4B), MOSAIC supports both zero-/few-shot prompting and lightweight fine-tuning, enabling deployment on consumer-grade GPUs. We evaluate MOSAIC across seven datasets in English, Spanish, French, and Danish, spanning multiple imaging modalities and label taxonomies. The model achieves a mean macro F1 score of 88 across five chest X-ray datasets, approaching or exceeding expert-level performance, while requiring only 24 GB of GPU memory. With data augmentation, as few as 80 annotated samples are sufficient to reach a weighted F1 score of 82 on Danish reports, compared to 86 with the full 1600-sample training set. MOSAIC offers a practical alternative to large or proprietary LLMs in clinical settings. Code and models are open-source. We invite the community to evaluate and extend MOSAIC on new languages, taxonomies, and modalities.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。