Building Efficient CNN Architectures for Histopathology Images Analysis: A Case-Study in Tumor-Infiltrating Lymphocytes Classification.
Building Efficient CNN Architectures for Histopathology Images Analysis: A Case-Study in Tumor-Infiltrating Lymphocytes Classification.
复制标题
DOI：
10.3389/fmed.2022.894430
复制
发表时间：
2022
影响因子：
3.9
通讯作者：
Teodoro, George
中科院分区：
文献类型：
--
作者：
Meirelles, Andre L. S.;Kurc, Tahsin;Kong, Jun;Ferreira, Renato;Saltz, Joel H.;Teodoro, George
深度学习方法在病理图像分析中表现出了卓越的性能，但它们的计算要求非常高。我们研究的目的是降低其计算成本，使其能够用于大型组织图像数据集。我们提出了一种称为网络自动归约（NAR）的方法，通过减少网络来简化卷积神经网络（CNN），以最大限度地减少进行预测的计算成本。NAR执行复合缩放，其中网络的宽度、深度和分辨率维度一起减小，以在所得到的简化网络中保持它们之间的平衡。我们比较我们的方法与一个国家的最先进的解决方案称为ResRep。该评估使用流行的CNN架构和识别组织图像中肿瘤浸润淋巴细胞分布的真实应用程序进行。实验结果表明，ResRep和NAR都能够生成简化的、更高效的ResNet50 V2版本。ResRep和NAR的简化版本分别需要比原始网络少1.32倍和3.26倍的浮点运算（FLOP），而不会损失分类能力（由曲线下面积（AUC）度量衡量）。当应用于更深、计算成本更高的网络时，Inception V4，NAR能够生成一个比原始版本低4倍的版本，同时具有相同的AUC性能。NAR能够大幅降低两种流行的CNN架构的执行成本，同时导致模型准确性的损失很小或没有损失。这种成本节约可以显着改善深度学习方法在数字病理学中的使用。它们可以使研究具有更大的组织图像数据集，并促进使用更便宜和更容易访问的图形处理单元（GPU），从而降低研究的计算成本。
Deep learning methods have demonstrated remarkable performance in pathology image analysis, but they are computationally very demanding. The aim of our study is to reduce their computational cost to enable their use with large tissue image datasets. We propose a method called Network Auto-Reduction (NAR) that simplifies a Convolutional Neural Network (CNN) by reducing the network to minimize the computational cost of doing a prediction. NAR performs a compound scaling in which the width, depth, and resolution dimensions of the network are reduced together to maintain a balance among them in the resulting simplified network. We compare our method with a state-of-the-art solution called ResRep. The evaluation is carried out with popular CNN architectures and a real-world application that identifies distributions of tumor-infiltrating lymphocytes in tissue images. The experimental results show that both ResRep and NAR are able to generate simplified, more efficient versions of ResNet50 V2. The simplified versions by ResRep and NAR require 1.32× and 3.26× fewer floating-point operations (FLOPs), respectively, than the original network without a loss in classification power as measured by the Area under the Curve (AUC) metric. When applied to a deeper and more computationally expensive network, Inception V4, NAR is able to generate a version that requires 4× lower than the original version with the same AUC performance. NAR is able to achieve substantial reductions in the execution cost of two popular CNN architectures, while resulting in small or no loss in model accuracy. Such cost savings can significantly improve the use of deep learning methods in digital pathology. They can enable studies with larger tissue image datasets and facilitate the use of less expensive and more accessible graphics processing units (GPUs), thus reducing the computing costs of a study.
登录
查看更多内容
影响因子：
7.7
作者：
Goyal M;Knackstedt T;Yan S;Hassanpour S
通讯作者：
Hassanpour S
影响因子：
6
作者：
Komura D;Ishikawa S
通讯作者：
Ishikawa S
影响因子：
3.9
作者：
Binder, Thomas;Tantaoui, El Mehdi;Gabrani, Maria
通讯作者：
Gabrani, Maria
通过多个实例学习的组织病理学图像分类和本地化的多分辨率模型。
影响因子：
7.7
作者：
Li J;Li W;Sisk A;Ye H;Wallace WD;Speier W;Arnold CW
通讯作者：
Arnold CW
影响因子：
7.7
作者：
George, Kalpana;Faziludeen, Shameer;Joseph, Paul K.
通讯作者：
Joseph, Paul K.