Interpretable medical image Visual Question Answering via multi-modal relationship graph learning

人工智能 情态动词 计算机科学 答疑 图形 水准点(测量) 任务(项目管理) 情报检索 机器学习 自然语言处理 理论计算机科学 经济 化学 管理 高分子化学 地理 大地测量学
作者
Xinyue Hu,Lin Gu,Kazuma Kobayashi,Liangchen Liu,Mengliang Zhang,Tatsuya Harada,Ronald M. Summers,Yingying Zhu
出处
期刊:Medical Image Analysis [Elsevier BV]
卷期号:97: 103279-103279 被引量:28
标识
DOI:10.1016/j.media.2024.103279
摘要

Medical Visual Question Answering (VQA) is an important task in medical multi-modal Large Language Models (LLMs), aiming to answer clinically relevant questions regarding input medical images. This technique has the potential to improve the efficiency of medical professionals while relieving the burden on the public health system, particularly in resource-poor countries. However, existing medical VQA datasets are small and only contain simple questions (equivalent to classification tasks), which lack semantic reasoning and clinical knowledge. Our previous work proposed a clinical knowledge-driven image difference VQA benchmark using a rule-based approach (Hu et al., 2023). However, given the same breadth of information coverage, the rule-based approach shows an 85% error rate on extracted labels. We trained an LLM method to extract labels with 62% increased accuracy. We also comprehensively evaluated our labels with 2 clinical experts on 100 samples to help us fine-tune the LLM. Based on the trained LLM model, we proposed a large-scale medical VQA dataset, Medical-CXR-VQA, using LLMs focused on chest X-ray images. The questions involved detailed information, such as abnormalities, locations, levels, and types. Based on this dataset, we proposed a novel VQA method by constructing three different relationship graphs: spatial relationships, semantic relationships, and implicit relationship graphs on the image regions, questions, and semantic labels. We leveraged graph attention to learn the logical reasoning paths for different questions. These learned graph VQA reasoning paths can be further used for LLM prompt engineering and chain-of-thought, which are crucial for further fine-tuning and training multi-modal large language models. Moreover, we demonstrate that our approach has the qualities of evidence and faithfulness, which are crucial in the clinical field. The code and the dataset is available at https://github.com/Holipori/Medical-CXR-VQA.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
刚刚
大月儿完成签到 ,获得积分10
刚刚
所所的应助被快乐友儿采纳,获得10
1秒前
LiKe完成签到 ,获得积分10
1秒前
赘婿的应助被daigang采纳,获得30
1秒前
汉堡包的应助被HU采纳,获得10
2秒前
姜汁完成签到 ,获得积分10
2秒前
二条完成签到,获得积分10
2秒前
2秒前
echo发布了新的文献求助10
2秒前
咿呀咿呀哟的应助被ypz采纳,获得10
3秒前
急聘行完成签到 ,获得积分10
3秒前
leotao发布了新的文献求助10
3秒前
脑洞疼的应助被安年采纳,获得10
3秒前
云宇完成签到 ,获得积分10
3秒前
小贾完成签到 ,获得积分10
4秒前
Yurith完成签到,获得积分10
4秒前
搜集达人的应助被顶刊相见采纳,获得10
4秒前
4秒前
星空发布了新的文献求助10
4秒前
一一完成签到,获得积分10
4秒前
初景发布了新的文献求助10
5秒前
一只盒子完成签到,获得积分10
5秒前
Akim的应助被风趣的水绿采纳,获得10
5秒前
自然菠萝完成签到,获得积分10
5秒前
乐乐完成签到,获得积分10
5秒前
hebnkygzs完成签到 ,获得积分10
5秒前
优雅沛文完成签到 ,获得积分10
5秒前
6秒前
CC66发布了新的文献求助10
6秒前
东山小红发布了新的文献求助10
6秒前
7秒前
7秒前
Firsterchao的应助被lufang采纳,获得10
8秒前
蜗牛完成签到,获得积分10
8秒前
violet完成签到,获得积分10
8秒前
柔弱的冰兰完成签到,获得积分20
9秒前
李健的应助被顺利的莺采纳,获得10
9秒前
YOUNG929的应助被快乐友儿采纳,获得10
10秒前
10秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
John McDowells Bildungstheorie 800
Social Psychology in the Real World 600
八旗制度の研究 600
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8180488
求助须知:如何正确求助?哪些是违规求助? 9570437
关于积分的说明 21812406
捐赠科研通 7776516
什么是DOI,文献DOI怎么找? 3433277
关于科研通互助平台的介绍 2557516
邀请新用户注册赠送积分活动 2479162