OmniBrainBench：面向多阶段临床任务的脑影像分析综合多模态基准
OmniBrainBench: A Comprehensive Multimodal Benchmark for Brain Imaging Analysis Across Multi-stage Clinical Tasks
为视觉语言模型在专业医学域提供覆盖15种脑成像模态、含封闭式与开放式问答的多阶段临床评测协议，可用于诊断MLLM在视觉到临床推理链路上的短板，并作为医学多模态模型选型与改进的对照基准。
原文依据 we introduce OmniBrainBench, the first comprehensive multimodal VQA benchmark specifically designed to assess the multimodal comprehension capabilities of MLLMs in brain imaging analysis
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论脑影像分析对脑部疾病的诊断与治疗至关重要，多模态大语言模型（MLLMs）正日益为其提供支持。然而，现有的脑影像视觉问答（VQA）基准要么覆盖的成像模态数量有限，要么局限于粗粒度的病理描述，难以在完整临床流程上对 MLLMs 进行全面评估。为此，我们提出 OmniBrainBench，这是首个专门用于评估 MLLMs 在脑影像分析中多模态理解能力的综合性多模态 VQA 基准，同时支持封闭式与开放式评测。OmniBrainBench 涵盖从 30 个经核实的医学来源收集的 15 种不同脑成像模态，共得到 9,527 个经校验的 VQA 问答对与 31,706 张图像。它模拟临床工作流程，包含 15 项多阶段临床任务，并由一位专业放射科医师严格校验。对 24 个最优模型的评估——包括开源通用、医学专用与闭源 MLLMs——凸显了 OmniBrainBench 带来的重大挑战。实验表明，Gemini-2.5-Pro（66.58%）等闭源 MLLMs 优于开源与医学专用 MLLMs，但仍远落后于医师（91.35%）；医学专用 MLLMs 在封闭式与开放式 VQA 上表现出较大波动。开源通用 MLLMs 总体落后，但在特定任务上表现突出；所有 MLLMs 在复杂的术前推理上均表现不足，揭示出关键的视觉到临床之间的差距。OmniBrainBench 为脑影像分析中的 MLLMs 评估建立了新标准，凸显了其与医师之间的差距。我们已在链接处公开基准与代码。
英文原文摘要 1723 字符
Brain imaging analysis is crucial for diagnosing and treating brain disorders, and multimodal large language models (MLLMs) are increasingly supporting it. However, current brain imaging visual question-answering (VQA) benchmarks either cover a limited number of imaging modalities or are restricted to coarse-grained pathological descriptions, hindering a comprehensive assessment of MLLMs across the full clinical continuum. To address these, we introduce OmniBrainBench, the first comprehensive multimodal VQA benchmark specifically designed to assess the multimodal comprehension capabilities of MLLMs in brain imaging analysis with closed- and open-ended evaluations. OmniBrainBench comprises 15 distinct brain imaging modalities collected from 30 verified medical sources, yielding 9,527 validated VQA pairs and 31,706 images. It simulates clinical workflows and encompasses 15 multi-stage clinical tasks rigorously validated by a professional radiologist. Evaluations of 24 state-of-the-art models, including open-source general-purpose, medical, and proprietary MLLMs, highlight the substantial challenges posed by OmniBrainBench. Experiments reveal that proprietary MLLMs like Gemini-2.5-Pro (66.58%) outperform open-source and medical MLLMs yet lag far behind physicians (91.35%), while medical MLLMs show wide variance in closed- and open-ended VQA. Open-source general-purpose MLLMs generally trail but excel in specific tasks, and all MLLMs fall short in complex preoperative reasoning, revealing a critical visual-to-clinical gap. OmniBrainBench establishes a new standard to assess MLLMs in brain imaging analysis, highlighting the gaps against physicians. We publicly release our benchmark and code at link.