Evaluation of Convolutional Neural Networks for Search in 1/f 2.8 Filtered Noise and Digital Breast Tomosynthesis Phantoms.
Evaluation of Convolutional Neural Networks for Search in 1/f 2.8 Filtered Noise and Digital Breast Tomosynthesis Phantoms.
复制标题
用于 1/f 2.8 过滤噪声和数字乳房断层合成模型搜索的卷积神经网络评估。
DOI：
10.1117/12.2549362
复制
发表时间：
2020
期刊：
影响因子：
--
通讯作者：
Eckstein,MiguelP
中科院分区：
文献类型：
--
作者：
Jonnalagadda,Aditya;Lago,MiguelA;Barufaldi,Bruno;Bakic,PredragR;Abbey,CraigK;Maidment,AndrewD;Eckstein,MiguelP
随着强大的卷积神经网络（CNN）的出现，最近的研究已经将神经网络的早期应用扩展到成像任务，从而使CNN成为评估医学图像质量的潜在新工具。在这里，我们将CNN与搜索任务中的模型观察者进行比较，以寻找嵌入在滤波噪声和数字乳腺断层合成（DBT）虚拟体模的单个切片中的两个可能信号（模拟肿块和较小的模拟微钙化）。对于过滤噪声的情况，我们展示了CNN如何近似搜索任务的理想观察者，微钙化的统计效率为0.77，肿块的统计效率为0.78。对于搜索DBT幻影在单个切片，我们表明，一个mixelized霍特林观察员（CHO）的性能受到影响，假阳性相关的解剖变异和结果低于人类观察员的性能检测精度。相比之下，CNN学习识别和折扣背景，并实现与人类观察者相当的性能和上级于模型观察者的性能（微钙化的比例校正：CNN = 0.96;人类= 0.98; CHO = 0.84;质量的比例校正：CNN = 0.98;人类= 0.83; CHO = 0.51）。总之，我们的研究结果通过在复杂的搜索任务中对人类和模型观察者的性能进行基准测试，为CNN方法提供了重要的评估。
With the advent of powerful convolutional neural networks (CNNs), recent studies have extended early applications of neural networks to imaging tasks thus making CNNs a potential new tool for assessing medical image quality. Here, we compare a CNN to model observers in a search task for two possible signals (a simulated mass and a smaller simulated micro-calcification) embedded in filtered noise and single slices of Digital Breast Tomosynthesis (DBT) virtual phantoms. For the case of the filtered noise, we show how a CNN can approximate the ideal observer for a search task, achieving a statistical efficiency of 0.77 for the microcalcification and 0.78 for the mass. For search in single slices of DBT phantoms, we show that a Channelized Hotelling Observer (CHO) performance is affected detrimentally by false positives related to anatomic variations and results in detection accuracy below human observer performance. In contrast, the CNN learns to identify and discount the backgrounds, and achieves performance comparable to that of human observer and superior to model observers (Proportion Correct for the microcalcification: CNN = 0.96; Humans = 0.98; CHO = 0.84; Proportion Correct for the mass: CNN = 0.98; Humans = 0.83; CHO = 0.51). Together, our results provide an important evaluation of CNN methods by benchmarking their performance against human and model observers in complex search tasks.