Article Figures & Data
Figures
- FIGURE 1.
CNN architectures used to learn μ-CT from λ-MLAA and μ-MLAA. (A) CAE. (B) Unet. (C) Hybrid network of CAE and Unet. Green and red vertical strips at far left indicate inputs to CNN, and red stripes at right indicate output. Each box represents multichannel feature map. Number of feature maps and dimension of each feature map are denoted on interior and bottom of box. Data flow is left to right through contracting path to capture context and symmetric expanding path to recover image. Arrows stand for copying feature maps, and sky-blue boxes are copied feature map.
- FIGURE 2.
Flow chart of image analysis. For comparison, emission PET sinogram was reconstructed using μ-maps obtained using MLAA before (μ-MLAA) and after (μ-CAE, μ-Unet, and μ-Hybrid) applying deep CNNs and ground truth μ-CT. TOF = time of flight.
- FIGURE 3.
Comparison of CNN outputs (μ-CAE, μ-Unet, and μ-Hybrid) to μ-MLAA and μ-CT. Red and yellow arrows indicate, respectively, crosstalk artifacts and bone estimation error shown in μ-MLAA.
- FIGURE 4.
Root-mean square errors (RMSE) relative to μ-CT plotted across slice axial location (average of 40 test sets).
- FIGURE 5.
Percentage error map of spatially normalized activity distribution (average of 40 test sets).
- FIGURE 6.
Percentage error in activity (A) and binding ratio (B) estimation relative to ground truth (OSEM with μ-CT). Each horizontal bar and vertical box indicates median and SD, respectively. In B, specific and nonspecific regions for binding ratio calculation are indicated as “specific (nonspecific).”
Tables
Whole head Cranial region Method Bone Air Bone Air MLAA 0.374 ± 0.058 0.317 ± 0.070 0.399 ± 0.063 0.426 ± 0.062 CNN (CAE) 0.717 ± 0.047 0.513 ± 0.057 0.747 ± 0.047 0.523 ± 0.063 CNN (Unet) 0.787 ± 0.042 0.575 ± 0.047 0.801 ± 0.043 0.580 ± 0.053 CNN (Hybrid) 0.794 ± 0.037 0.718 ± 0.048 0.810 ± 0.038 0.738 ± 0.044 Data are mean ± SD. Results of analysis of variation and post hoc tests are shown in Supplemental Figure 3.
Jump to section
Related Articles
Cited By...
- Improving 18F-FDG PET Quantification Through a Spatial Normalization Method
- Nuclear Medicine and Artificial Intelligence: Best Practices for Algorithm Development
- PET/MRI, Part 2: Technologic Principles
- Machine Learning in Nuclear Medicine: Part 2--Neural Networks and Clinical Aspects
- Denoising of Scintillation Camera Images Using a Deep Convolutional Neural Network: A Monte Carlo Simulation Approach
- Intelligent Imaging: Artificial Intelligence Augmented Nuclear Medicine
- Generation of PET Attenuation Map for Whole-Body Time-of-Flight 18F-FDG PET/MRI Using a Deep Neural Network Trained with Simultaneously Reconstructed Activity and Attenuation Maps