Cognitive Science to Radiology: Using EEG and Eye-tracking to Determine Why, How, and When Novices and Radiologists Miss Signs of Breast Cancer in Multiple-abnormality Mammography and Tomosynthesis

从认知科学到放射学：使用脑电图和眼动追踪来确定新手和放射科医生在多异常乳房 X 线摄影和断层合成中错过乳腺癌迹象的原因、方式和时间

• 批准号: 10525777
• 负责人: Stephen Adamo
• 金额: $ 12.53万
• 依托单位: UNIVERSITY OF CENTRAL FLORIDA
• 依托单位国家: 美国
• 财政年份: 2022
• 资助国家: 美国
• 起止时间: 2022-09-01 至 2024-08-30
• 项目状态: 已结题
• 来源: https://reporter.nih.gov/project-details/10525777
• 关键词: 3-Dimensional; Address; Affect; Attention; Breast Cancer Detection; Cancer Detection; Cognitive; Cognitive Science; Data; Detection; Digital Breast Tomosynthesis; Digital Mammography; Educational workshop; Electroencephalogram; Goals; Grant; Image; Imaging technology; Institution; Knowledge; Life; Malignant Neoplasms; Mammography; Medical Imaging; Memory; Mentors; Methodology; Methods; Mission; Modality; Multiple Abnormalities; Normalcy; Outcome; Public Health; Radiology Specialty; Research; Research Personnel; Research Technics; Screening for cancer; Screening procedure; Short-Term Memory; Techniques; Testing; Three-Dimensional Image; Three-Dimensional Imaging; Time; Training; United States; United States National Institutes of Health; Visual; Woman; Work; attentional bias; base; breast imaging; cancer diagnosis; cancer imaging; improved; innovation; malignant breast neoplasm; meetings; mental representation; neuromechanism; novel; oculomotor; radiological imaging; radiologist; relating to nervous system; satisfaction; tomosynthesis; virtual; visual tracking

Satisfaction of Search (SOS), where an abnormality is missed after a previous abnormality was found, is a pervasive problem across different types of images, yet their impact on breast cancer detection is un- known. Radiology and cognitive science have independently investigated why, how, and when SOS oc- curs. The objectives of this grant are to bridge cognitive science research techniques with radiology to de- termine: 1) the neural underpinnings of SOS, 2) the extent to which SOS affects breast cancer detection, 3) the extent to which advancements in imaging technology alleviates SOS, and 4) how expertise affects SOS error rates. My central hypothesis is SOS contributes to breast cancer misses because a found ab- normality becomes an attentional template. An attentional template is known in cognitive science as an enhanced representation of a target maintained within memory and biases attention in search. The ra- tionale for this project is when a first target is detected, it becomes an attentional template that utilizes and biases attention necessary for detecting subsequent targets making searchers prone to SOS. However, SOS may be mitigated: 1) by changing search strategies to counteract a first target becoming an atten- tional template, 2) with advances in imaging technology, and 3) with expertise in breast cancer detection. These influences need to be investigated to determine the extent to which they can alleviate SOS. This work will pursue four specific aims to test the central hypothesis: 1) identify a mechanistic account for SOS, 2) determine the extent to which SOS accounts for breast cancer misses, 3) compare SOS between mammography and tomosynthesis, and 4) compare SOS error rates between different levels of expertise. The research proposed is innovative because it will: 1) identify a mechanism responsible for SOS; 2) use EEG and eye-tracking to study the neural basis of SOS; 3) use realistic, virtual breast images to investi- gate the extent to which SOS affects breast cancer detection; 4) determine the extent to which the SOS results from simplified search displays replicate in realistic-looking medical images; 5) be the first time SOS is investigated with 3D imaging; and 6) determine the extent to which SOS rates vary by expertise in breast cancer detection. The proposed research is significant because: 1) it will develop a neural under- standing of how SOS contributes to cancer detection; 2) determine how SOS varies across different search modalities and by expertise; and 3) ultimately improve breast cancer detection by determining how these influences impact SOS rates. To complete the proposed research, in pursuit of his long-term goal of becoming an independent researcher at a Tier-I institution, the PI will train using EEG and eye-tracking. His training will include auditing graduate courses, attending workshops, weekly meetings/training with his mentors, and bi-annual assessments of his EEG/eye-tracking expertise with his mentors.

搜索满意度（SOS），即在发现先前的异常后漏掉该异常，是一种 不同类型图像中普遍存在的问题，但它们对乳腺癌检测的影响却是未知的。 已知。放射学和认知科学独立研究了 SOS 发生的原因、方式和时间。 诅咒。这笔赠款的目的是将认知科学研究技术与放射学联系起来，以解决 终点：1) SOS 的神经基础，2) SOS 影响乳腺癌检测的程度， 3) 成像技术的进步在多大程度上缓解了 SOS，以及 4) 专业知识如何影响 SOS 错误率。我的中心假设是 SOS 导致乳腺癌漏诊，因为发现了 常态成为注意力模板。注意力模板在认知科学中被称为 增强记忆中目标的表征并在搜索中偏向注意力。拉- 该项目的要点是，当检测到第一个目标时，它就成为一个注意力模板，利用和 会偏向检测后续目标所需的注意力，使搜索者容易发出求救信号。然而， SOS 可以通过以下方式缓解：1）通过改变搜索策略来抵消第一个目标成为关注者 模板，2) 成像技术的进步，3) 乳腺癌检测方面的专业知识。 需要对这些影响进行研究，以确定它们可以在多大程度上缓解 SOS。这 工作将追求四个具体目标来检验中心假设：1）确定机械解释 SOS，2) 确定 SOS 对乳腺癌漏诊的影响程度，3) 比较 SOS 乳房 X 光检查和断层合成，4) 比较不同专业水平之间的 SOS 错误率。 所提出的研究具有创新性，因为它将：1）确定负责 SOS 的机制； 2）使用 脑电图和眼动追踪研究 SOS 的神经基础； 3）使用真实的、虚拟的乳房图像来调查 确定 SOS 对乳腺癌检测的影响程度； 4）确定SOS的程度 简化搜索显示的结果复制在逼真的医学图像中； 5）第一次 通过 3D 成像研究 SOS； 6) 确定 SOS 率因专业知识而变化的程度 乳腺癌检测。拟议的研究意义重大，因为：1）它将开发一种神经底层- SOS 如何促进癌症检测的地位； 2) 确定 SOS 在不同的环境中如何变化 搜索方式和专业知识； 3) 通过确定如何最终改善乳腺癌检测 这些因素都会影响 SOS 率。完成拟议的研究，以追求他的长期目标 成为一级机构的独立研究员后，PI 将使用脑电图和眼动追踪进行培训。 他的培训将包括旁听研究生课程、参加研讨会、与他的导师每周举行会议/培训 导师，以及与导师一起对他的脑电图/眼动追踪专业知识进行每两年一次的评估。