2.5D Deep Learning For CT Image Reconstruction Using A Multi-GPU Implementation
| Autor: | Charles A. Bouman, Amirkoushyar Ziabari, Ken David Sauer, Dong Hye Ye, Jean-Baptiste Thibault, Somesh Srivastava |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Radon transform
Computer science Image quality business.industry Deep learning Image and Video Processing (eess.IV) Reconstruction algorithm 02 engineering and technology Iterative reconstruction Electrical Engineering and Systems Science - Image and Video Processing Residual Convolutional neural network 030218 nuclear medicine & medical imaging 03 medical and health sciences 0302 clinical medicine Kernel (image processing) FOS: Electrical engineering electronic engineering information engineering 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Computer vision Artificial intelligence business |
| Zdroj: | ACSSC |
| DOI: | 10.1109/acssc.2018.8645364 |
| Popis: | While Model Based Iterative Reconstruction (MBIR) of CT scans has been shown to have better image quality than Filtered Back Projection (FBP), its use has been limited by its high computational cost. More recently, deep convolutional neural networks (CNN) have shown great promise in both denoising and reconstruction applications. In this research, we propose a fast reconstruction algorithm, which we call Deep Learning MBIR (DL-MBIR), for approximating MBIR using a deep residual neural network. The DL-MBIR method is trained to produce reconstructions that approximate true MBIR images using a 16 layer residual convolutional neural network implemented on multiple GPUs using Google Tensorflow. In addition, we propose 2D, 2.5D and 3D variations on the DL-MBIR method and show that the 2.5D method achieves similar quality to the fully 3D method, but with reduced computational cost. IEEE Asilomar conference on signals systems and computers, 2018 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|