• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

创建并验证一个带有眼动追踪和报告口述功能的胸部 X 射线数据集，用于人工智能开发。

Creation and validation of a chest X-ray dataset with eye-tracking and report dictation for AI development.

机构信息

IBM Research, Almaden Research Center, San Jose, CA, 95120, USA.

Department of Computer Science, Virginia Tech, Blacksburg, VA, 24061, USA.

出版信息

Sci Data. 2021 Mar 25;8(1):92. doi: 10.1038/s41597-021-00863-5.

DOI:10.1038/s41597-021-00863-5
PMID:33767191
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC7994908/
Abstract

We developed a rich dataset of Chest X-Ray (CXR) images to assist investigators in artificial intelligence. The data were collected using an eye-tracking system while a radiologist reviewed and reported on 1,083 CXR images. The dataset contains the following aligned data: CXR image, transcribed radiology report text, radiologist's dictation audio and eye gaze coordinates data. We hope this dataset can contribute to various areas of research particularly towards explainable and multimodal deep learning/machine learning methods. Furthermore, investigators in disease classification and localization, automated radiology report generation, and human-machine interaction can benefit from these data. We report deep learning experiments that utilize the attention maps produced by the eye gaze dataset to show the potential utility of this dataset.

摘要

我们开发了一个丰富的 Chest X-Ray（CXR）图像数据集，以协助研究人员进行人工智能研究。这些数据是使用眼动追踪系统收集的，当时一名放射科医生对 1083 张 CXR 图像进行了审查和报告。该数据集包含以下对齐数据：CXR 图像、转录的放射学报告文本、放射科医生的口述音频和眼动坐标数据。我们希望这个数据集可以为各个研究领域做出贡献，特别是在可解释性和多模态深度学习/机器学习方法方面。此外，疾病分类和定位、自动化放射学报告生成以及人机交互方面的研究人员也可以从这些数据中受益。我们报告了利用眼动数据集生成的注意力图进行的深度学习实验，以展示该数据集的潜在用途。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/b7543733736f/41597_2021_863_Fig17_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/c2ca1d209352/41597_2021_863_Fig1_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/00abe7ee541b/41597_2021_863_Fig2_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/38c739da1805/41597_2021_863_Fig3_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/57b53c8d85ea/41597_2021_863_Fig4_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/0aceae5b0a2d/41597_2021_863_Fig5_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/c3e3c382a4d2/41597_2021_863_Fig6_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/195b80eb3660/41597_2021_863_Fig7_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/86535b18597a/41597_2021_863_Fig8_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/7dc062352dae/41597_2021_863_Fig9_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/7bb40a6db58f/41597_2021_863_Fig10_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/4f837b5951f5/41597_2021_863_Fig11_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/743f7cb3db32/41597_2021_863_Fig12_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/8af36f812da5/41597_2021_863_Fig13_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/5de27b5a4d06/41597_2021_863_Fig14_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/883ccc695ab5/41597_2021_863_Fig15_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/55e8a3a92dbf/41597_2021_863_Fig16_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/b7543733736f/41597_2021_863_Fig17_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/c2ca1d209352/41597_2021_863_Fig1_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/00abe7ee541b/41597_2021_863_Fig2_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/38c739da1805/41597_2021_863_Fig3_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/57b53c8d85ea/41597_2021_863_Fig4_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/0aceae5b0a2d/41597_2021_863_Fig5_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/c3e3c382a4d2/41597_2021_863_Fig6_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/195b80eb3660/41597_2021_863_Fig7_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/86535b18597a/41597_2021_863_Fig8_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/7dc062352dae/41597_2021_863_Fig9_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/7bb40a6db58f/41597_2021_863_Fig10_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/4f837b5951f5/41597_2021_863_Fig11_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/743f7cb3db32/41597_2021_863_Fig12_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/8af36f812da5/41597_2021_863_Fig13_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/5de27b5a4d06/41597_2021_863_Fig14_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/883ccc695ab5/41597_2021_863_Fig15_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/55e8a3a92dbf/41597_2021_863_Fig16_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/d695/7994908/b7543733736f/41597_2021_863_Fig17_HTML.jpg

相似文献

1
Creation and validation of a chest X-ray dataset with eye-tracking and report dictation for AI development.创建并验证一个带有眼动追踪和报告口述功能的胸部 X 射线数据集，用于人工智能开发。
Sci Data. 2021 Mar 25;8(1):92. doi: 10.1038/s41597-021-00863-5.
2
REFLACX, a dataset of reports and eye-tracking data for localization of abnormalities in chest x-rays.REFLACX，一个包含报告和眼动数据的数据集，用于定位胸部 X 光片中的异常。
Sci Data. 2022 Jun 18;9(1):350. doi: 10.1038/s41597-022-01441-z.
3
Chest X-ray image phase features for improved diagnosis of COVID-19 using convolutional neural network.基于卷积神经网络的胸部 X 射线图像相位特征提高 COVID-19 诊断性能
Int J Comput Assist Radiol Surg. 2021 Feb;16(2):197-206. doi: 10.1007/s11548-020-02305-w. Epub 2021 Jan 9.
4
Shedding light on ai in radiology: A systematic review and taxonomy of eye gaze-driven interpretability in deep learning.解读放射学中的人工智能：深度学习中眼动驱动可解释性的系统综述与分类法
Eur J Radiol. 2024 Mar;172:111341. doi: 10.1016/j.ejrad.2024.111341. Epub 2024 Feb 1.
5
Diagnostic performance of artificial intelligence model for pneumonia from chest radiography.基于胸部 X 光的人工智能肺炎诊断模型的性能。
PLoS One. 2021 Apr 15;16(4):e0249399. doi: 10.1371/journal.pone.0249399. eCollection 2021.
6
Deep Learning Models to Predict Fatal Pneumonia Using Chest X-Ray Images.深度学习模型利用胸部 X 光图像预测致命性肺炎。
Can Respir J. 2022 Nov 24;2022:8026580. doi: 10.1155/2022/8026580. eCollection 2022.
7
CheXGAT: A disease correlation-aware network for thorax disease diagnosis from chest X-ray images.CheXGAT：一种用于从胸部X光图像进行胸部疾病诊断的疾病关联感知网络。
Artif Intell Med. 2022 Oct;132:102382. doi: 10.1016/j.artmed.2022.102382. Epub 2022 Aug 27.
8
Learning to diagnose common thorax diseases on chest radiographs from radiology reports in Vietnamese.从越南放射科的放射报告中学习诊断常见胸部疾病的胸片。
PLoS One. 2022 Oct 31;17(10):e0276545. doi: 10.1371/journal.pone.0276545. eCollection 2022.
9
Improving disease classification performance and explainability of deep learning models in radiology with heatmap generators.使用热图生成器提高放射学中深度学习模型的疾病分类性能和可解释性。
Front Radiol. 2022 Oct 11;2:991683. doi: 10.3389/fradi.2022.991683. eCollection 2022.
10
A disentangled generative model for disease decomposition in chest X-rays via normal image synthesis.通过正常图像合成对胸部 X 光片中的疾病进行分解的解缠生成模型。
Med Image Anal. 2021 Jan;67:101839. doi: 10.1016/j.media.2020.101839. Epub 2020 Oct 7.

引用本文的文献

1
Joint enhancement of automatic chest x-ray diagnosis and radiological gaze prediction with multistage cooperative learning.通过多阶段协作学习联合增强自动胸部X光诊断和放射学注视预测
Med Phys. 2025 Jul;52(7):e17977. doi: 10.1002/mp.17977.
2
Enhancing colorectal polyp classification using gaze-based attention networks.使用基于注视的注意力网络增强结直肠息肉分类
PeerJ Comput Sci. 2025 Mar 25;11:e2780. doi: 10.7717/peerj-cs.2780. eCollection 2025.
3
DECODING RADIOLOGISTS' INTENTIONS: A NOVEL SYSTEM FOR ACCURATE REGION IDENTIFICATION IN CHEST X-RAY IMAGE ANALYSIS.

本文引用的文献

1
AI Accelerated Human-in-the-loop Structuring of Radiology Reports.人工智能加速放射科报告的人机交互结构。
AMIA Annu Symp Proc. 2021 Jan 25;2020:1305-1314. eCollection 2020.
2
Automatic Lung Nodule Detection Combined With Gaze Information Improves Radiologists' Screening Performance.自动肺结节检测结合注视信息可提高放射科医生的筛查性能。
IEEE J Biomed Health Inform. 2020 Oct;24(10):2894-2901. doi: 10.1109/JBHI.2020.2976150. Epub 2020 Feb 24.
3
Assessing Radiology Research on Artificial Intelligence: A Brief Guide for Authors, Reviewers, and Readers-From the Editorial Board.
解读放射科医生的意图：一种用于胸部X光图像分析中准确区域识别的新型系统。
Proc IEEE Int Symp Biomed Imaging. 2024 May;2024. doi: 10.1109/isbi56570.2024.10635322. Epub 2024 Aug 22.
4
Diagnosis and Treatment of Primary Tracheobronchial Tumors.原发性气管支气管肿瘤的诊断与治疗
Cancer Med. 2025 May;14(9):e70893. doi: 10.1002/cam4.70893.
5
Modeling radiologists' cognitive processes using a digital gaze twin to enhance radiology training.使用数字凝视替身对放射科医生的认知过程进行建模以加强放射科培训。
Sci Rep. 2025 Apr 21;15(1):13685. doi: 10.1038/s41598-025-97935-y.
6
Multimodal contrastive learning for enhanced explainability in pediatric brain tumor molecular diagnosis.用于增强小儿脑肿瘤分子诊断可解释性的多模态对比学习
Sci Rep. 2025 Mar 30;15(1):10943. doi: 10.1038/s41598-025-94806-4.
7
Discrimination of Radiologists' Experience Level Using Eye-Tracking Technology and Machine Learning: Case Study.利用眼动追踪技术和机器学习辨别放射科医生的经验水平：案例研究
JMIR Form Res. 2025 Jan 22;9:e53928. doi: 10.2196/53928.
8
ItpCtrl-AI: End-to-end interpretable and controllable artificial intelligence by modeling radiologists' intentions.ItpCtrl-AI：通过对放射科医生的意图进行建模实现端到端可解释和可控的人工智能。
Artif Intell Med. 2025 Feb;160:103054. doi: 10.1016/j.artmed.2024.103054. Epub 2024 Dec 12.
9
Bridging human and machine intelligence: Reverse-engineering radiologist intentions for clinical trust and adoption.架起人类与机器智能之间的桥梁：逆向工程放射科医生建立临床信任和实现应用的意图。
Comput Struct Biotechnol J. 2024 Nov 8;24:711-723. doi: 10.1016/j.csbj.2024.11.012. eCollection 2024 Dec.
10
From explanation to intervention: Interactive knowledge extraction from Convolutional Neural Networks used in radiology.从解释到干预：从放射学中使用的卷积神经网络中交互式提取知识。
PLoS One. 2024 Apr 10;19(4):e0293967. doi: 10.1371/journal.pone.0293967. eCollection 2024.
评估人工智能放射学研究：给作者、审稿人和读者的简要指南——来自编辑委员会
Radiology. 2020 Mar;294(3):487-489. doi: 10.1148/radiol.2019192515. Epub 2019 Dec 31.
4
MIMIC-CXR, a de-identified publicly available database of chest radiographs with free-text reports.MIMIC-CXR，一个去标识化的、公开可用的、包含自由文本报告的胸部 X 光数据库。
Sci Data. 2019 Dec 12;6(1):317. doi: 10.1038/s41597-019-0322-0.
5
Analysis of Perceptual Expertise in Radiology - Current Knowledge and a New Perspective.放射学中的感知专业技能分析——当前知识与新视角
Front Hum Neurosci. 2019 Jun 25;13:213. doi: 10.3389/fnhum.2019.00213. eCollection 2019.
6
Eye Tracking for Deep Learning Segmentation Using Convolutional Neural Networks.基于卷积神经网络的深度学习分割的眼动追踪。
J Digit Imaging. 2019 Aug;32(4):597-604. doi: 10.1007/s10278-019-00220-4.
7
A collaborative computer aided diagnosis (C-CAD) system with eye-tracking, sparse attentional model, and deep learning.基于眼动追踪、稀疏注意力模型和深度学习的协同计算机辅助诊断 (C-CAD) 系统。
Med Image Anal. 2019 Jan;51:101-115. doi: 10.1016/j.media.2018.10.010. Epub 2018 Oct 28.
8
Modeling visual search behavior of breast radiologists using a deep convolution neural network.使用深度卷积神经网络对乳腺放射科医生的视觉搜索行为进行建模。
J Med Imaging (Bellingham). 2018 Jul;5(3):035502. doi: 10.1117/1.JMI.5.3.035502. Epub 2018 Aug 11.
9
How visual search relates to visual diagnostic performance: a narrative systematic review of eye-tracking research in radiology.视觉搜索与视觉诊断性能的关系：放射学中眼动追踪研究的叙述性系统综述
Adv Health Sci Educ Theory Pract. 2017 Aug;22(3):765-787. doi: 10.1007/s10459-016-9698-1. Epub 2016 Jul 19.
10
Investigating the link between radiologists' gaze, diagnostic decision, and image content.探讨放射科医生的注视、诊断决策与图像内容之间的关系。
J Am Med Inform Assoc. 2013 Nov-Dec;20(6):1067-75. doi: 10.1136/amiajnl-2012-001503. Epub 2013 Jun 20.