arXiv 2605.224142026-05-21cs.CVcs.AI

迈向具有空间定位病变证据的临床可解释性眼科VQA

Towards Clinically Interpretable Ophthalmic VQA via Spatially-Grounded Lesion Evidence

Department of Computer Science and Engineering, Southern University of Science and Technology（南方科技大学计算机科学与工程系）；The Hong Kong Polytechnic University（香港理工大学）；National University of Singapore（新加坡国立大学）；University of Washington（华盛顿大学）；Institute of High Performance Computing, Agency for Science, Technology and Research（科技研究局高性能计算研究所）

Xingyue Wang, Bo Liu, Meng Wang, Zhixuan Zhang, Chengcheng Zhu, Huazhu Fu, Jiang Liu

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出FundusGround基准，通过空间定位病变证据提升眼科VQA的临床可解释性，通过三阶段流程收集标注病变的视网膜影像，并评估多种视觉语言模型在答案准确性和病变层面推理上的表现。

摘要：

视觉问答（VQA）在临床支持中具有巨大潜力，特别是在眼科领域，视网膜彩色照相是诊断的关键。然而，眼科VQA基准主要强调答案准确性，忽视了临床可解释性所需的显式视觉证据。本文引入FundusGround，一个新的具有空间定位病变证据的临床可解释性眼科VQA基准。具体而言，我们提出一个三阶段流程，收集了10,719张带有15,595个图像级精细标注病变的视网膜影像。为确保解剖一致性和临床有效性，所有病变均通过早期治疗糖尿病视网膜病变研究（ETDRS）网格进行空间定位，从而标准化映射到九个具有临床意义的视网膜区域。基于此结构化的病变证据，生成了72,706个问题，涵盖四种格式：开放式、封闭式、单选和多选。我们进一步使用双指标（答案准确性和病变层面推理）评估多种通用和医学大型视觉语言模型。实验表明，整合病变层面的视觉证据能持续提高模型性能和透明度，突显了显式空间定位对于可靠和可解释性眼科VQA的必要性。

英文摘要：

Visual Question Answering (VQA) holds great promise for clinical support, particularly in ophthalmology, where retinal fundus photography is essential for diagnosis. However, ophthalmic VQA benchmarks primarily emphasize answer accuracy, neglecting the explicit visual evidence necessary for clinical interpretability. In this work, we introduce FundusGround, a new benchmark for clinically interpretable ophthalmic VQA with spatially-grounded lesion evidence. Specifically, we propose a three-stage pipeline that collects 10,719 fundus images with 15,595 image-level meticulously annotated lesions. To ensure anatomical consistency and clinical validity, all lesions are spatially localized using the Early Treatment Diabetic Retinopathy Study (ETDRS) grid, enabling standardized mapping to nine clinically meaningful retinal regions. Built upon this structured lesion evidence, 72,706 questions are then generated spanning four formats: open-ended, closed-ended, single-choice, and multiple-choice. We further benchmark multiple general- and medical- large vision-language models using dual metrics for answer accuracy and lesion-level reasoning. The experiments demonstrate that incorporating lesion-level visual evidence consistently improves model performance and transparency, highlighting the necessity of explicit spatial grounding for reliable and explainable ophthalmic VQA.

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。