Xing Xin, Liang Gongbo, Blanton Hunter, Rafique Muhammad Usman, Wang Chris, Lin Ai-Ling, Jacobs Nathan
University of Kentucky, Lexington KY 40506, USA.
Comput Vis ECCV. 2020 Aug;12535:355-364. doi: 10.1007/978-3-030-66415-2_23. Epub 2021 Jan 10.
We propose to apply a 2D CNN architecture to 3D MRI image Alzheimer's disease classification. Training a 3D convolutional neural network (CNN) is time-consuming and computationally expensive. We make use of approximate rank pooling to transform the 3D MRI image volume into a 2D image to use as input to a 2D CNN. We show our proposed CNN model achieves 9.5% better Alzheimer's disease classification accuracy than the baseline 3D models. We also show that our method allows for efficient training, requiring only 20% of the training time compared to 3D CNN models. The code is available online: https://github.com/UkyVision/alzheimer-project.
我们建议将二维卷积神经网络（2D CNN）架构应用于三维磁共振成像（3D MRI）图像的阿尔茨海默病分类。训练三维卷积神经网络（3D CNN）既耗时又耗费计算资源。我们利用近似秩池化将3D MRI图像体转换为二维图像，以用作2D CNN的输入。我们表明，我们提出的CNN模型在阿尔茨海默病分类准确率上比基线3D模型高9.5%。我们还表明，我们的方法能够实现高效训练，与3D CNN模型相比，所需训练时间仅为其20%。代码可在网上获取：https://github.com/UkyVision/alzheimer-project 。
