摘要：脑肿瘤诊断是一个时间敏感的过程，患者可能需要等待数周才能获得最终的病理报告。这一问题推动了能够基于多模态输入对肿瘤亚型进行分类的自动化系统的发展。本文介绍了 DS@GT ARC 团队参与 ImageCLEFmed MEDIQA-CORE 2026 Task~1——脑肿瘤亚型分类——的工作。该任务评估三项胶质瘤分类问题：Level-1 分子类型、LGG 与 HGG 分类，以及 WHO 分级。我们将预先提取的 MRI（NeuroVFM）和组织病理学（Prov-GigaPath）嵌入表示，与自由文本放射学报告相结合。团队探索了两种三模态融合架构、两种报告编码器（RadBERT 和 Llama-3.1-8B-Instruct），以及一个具有生物学动机的后处理阶段。在完全多模态条件下，我们取得了 0.801 的平均 macro-F1，超过主办方 0.796 的基线，并在代码通过验证的团队中排名第二。在模态缺失条件下进行的额外评估显示，这一优势高度依赖组织病理学模态的可用性；当存在模态缺失时，我们的系统表现落后于基线。我们的代码已发布于 GitHub：此 HTTPS 链接。

Abstract:Brain tumor diagnosis is a time-sensitive process in which patients may wait weeks for a finalized pathology report. This problem motivates automated systems that classify tumor subtype from multimodal inputs. This paper details the DS@GT ARC team's work for ImageCLEFmed MEDIQA-CORE 2026 Task~1, Brain Tumor Subtype Classification. The task evaluates three glioma classification problems: Level-1 Molecular Type, LGG vs HGG, and WHO Grade. We combine pre-extracted MRI (NeuroVFM) and histopathology (Prov-GigaPath) embeddings with free-text radiology reports. Our team explored two trimodal fusion architectures, two report encoders (RadBERT and Llama-3.1-8B-Instruct), and a biologically motivated post-processing stage. We achieve a mean macro-F1 of 0.801 under the Fully Multimodal condition, exceeding the organizers' baseline of 0.796 and ranking second among the teams whose code passed verification. Additional evaluation across modality-dropping conditions shows that this advantage depends heavily on the availability of the histopathology modality, and that our system falls behind the baseline when modalities are missing. Our code is available on GitHub at this https URL.