González Germán, Washko George R, Estépar Raúl San José
Sierra Research S.L. Avda Costa Blanca 132. Alicante. Spain.
Division of Pulmonary and Critical Care Medicine. Department of Medicine. Brigham and Women's Hospital. 75 Francis St, Boston, MA. USA.
Proc SPIE Int Soc Opt Eng. 2018 Feb;10574. doi: 10.1117/12.2293455. Epub 2018 Mar 2.
Biomarker computation using deep-learning often relies on a two-step process, where the deep learning algorithm segments the region of interest and then the biomarker is measured. We propose an alternative paradigm, where the biomarker is estimated directly using a regression network. We showcase this image-to-biomarker paradigm using two biomarkers: the estimation of bone mineral density (BMD) and the estimation of lung percentage of emphysema from CT scans.
We use a large database of 9,925 CT scans to train, validate and test the network for which reference standard BMD and percentage emphysema have been already computed. First, the 3D dataset is reduced to a set of canonical 2D slices where the organ of interest is visible (either spine for BMD or lungs for emphysema). This data reduction is performed using an automatic object detector. Second, The regression neural network is composed of three convolutional layers, followed by a fully connected and an output layer. The network is optimized using a momentum optimizer with an exponential decay rate, using the root mean squared error as cost function.
The Pearson correlation coefficients obtained against the reference standards are = 0.940 ( < 0.00001) and = 0.976 ( < 0.00001) for BMD and percentage emphysema respectively.
The deep-learning regression architecture can learn biomarkers from images directly, without indicating the structures of interest. This approach simplifies the development of biomarker extraction algorithms. The proposed data reduction based on object detectors conveys enough information to compute the biomarkers of interest.
使用深度学习进行生物标志物计算通常依赖于两步过程，即深度学习算法先分割感兴趣区域，然后测量生物标志物。我们提出了一种替代范式，即使用回归网络直接估计生物标志物。我们使用两种生物标志物展示这种图像到生物标志物的范式：从CT扫描中估计骨密度（BMD）和肺气肿的肺占比。
我们使用一个包含9925份CT扫描的大型数据库来训练、验证和测试网络，该数据库中已计算出参考标准BMD和肺气肿占比。首先，将3D数据集简化为一组感兴趣器官可见的规范2D切片（对于BMD是脊柱，对于肺气肿是肺）。这种数据简化使用自动目标检测器进行。其次，回归神经网络由三个卷积层组成，随后是一个全连接层和一个输出层。使用带有指数衰减率的动量优化器对网络进行优化，使用均方根误差作为代价函数。
相对于参考标准，BMD和肺气肿占比的皮尔逊相关系数分别为 = 0.940（< 0.00001）和 = 0.976（< 0.00001）。
深度学习回归架构可以直接从图像中学习生物标志物，而无需指明感兴趣的结构。这种方法简化了生物标志物提取算法的开发。所提出的基于目标检测器的数据简化传达了足够的信息来计算感兴趣的生物标志物。
