Post-acquisition mask misalignment correction for edge illumination x-ray phase contrast imaging.
D ShoukrounAdam DohertyMarco EndrizziD BatePaul FrommeAlessandro OlivoPublished in: The Review of scientific instruments (2022)
Edge illumination x-ray phase contrast imaging uses a set of apertured masks to translate phase effects into variation of detected intensity. While the system is relatively robust against misalignment, mask movement during acquisition can lead to gradient artifacts. A method has been developed to correct the images by quantifying the misalignment post-acquisition and implementing correction maps to remove the gradient artifact. Images of a woven carbon fiber composite plate containing porosity were used as examples to demonstrate the image correction process. The gradient formed during image acquisition was removed without affecting the image quality, and results were subsequently used for quantification of porosity, indicating that the gradient correction did not affect the quantitative content of the images.
Keyphrases
- deep learning
- image quality
- high resolution
- dual energy
- convolutional neural network
- computed tomography
- magnetic resonance
- optical coherence tomography
- contrast enhanced
- machine learning
- high intensity
- quality improvement
- magnetic resonance imaging
- positive airway pressure
- obstructive sleep apnea
- fluorescence imaging