Analyzer-Less X-Ray Interferometry with Super-Resolution Methods πŸ“

Author: Joyoni Dey, Hunter Cole Meyer, Murtuza Taqi πŸ‘¨β€πŸ”¬

Affiliation: Louisiana State University 🌍

Abstract:

Purpose: We propose using super-resolution methods for X-ray grating interferometry without an analyzer with detectors that fail to meet the Nyquist sampling rate needed for traditional image recovery algorithms. This method enables the Talbot-Lau interferometer without the X-ray absorbing analyzer and allows for higher autocorrelation lengths for the analyzer-less Modulated Phase Grating Interferometer. This will lead to reduced X-ray dose by a factor of 2 and higher autocorrelation lengths than previously accessible.
Methods: We recover the projection signals in three stages. In Stage I, the subpixel phase steps (each aliased) are stacked and rasterized to recover the sampling. This recovers the sampling to phase-step resolution. In Stage II, the visibility loss for the reference signal is iteratively recovered using quasi-newton minimizing of sum-square error. In Stage III, we recover object parameters in an iterative gradient descent maximization of the likelihood function by iteratively applying the attenuation, differential phase, and the small angle scatter on the recovered reference (no object) signal (Stage II). The small angle scatter is modeled as a visibility drop for the first harmonic.
This recovery method is tested by simulating normal and diseased lung tissue with a tumor. Eight cases were testedβ€”two different detectors (30, 50 micro-m), two different phase step resolutions, and two different sets of noise were tested on a simulated sample with small-angle scatter signals for lungs with tumors.
Results: The three parameters were recovered well for all 8 cases. The attenuation had the least error < 1%. The small angle scatter visibility had higher RMSE values at 4-14%.
Conclusion: We present a novel approach to perform analyzer-less interferometry using super-resolution methods and iterative reconstruction to recover the attenuation, differential phase, and dark-field images. For the next steps, we will image mice lungs with different interferometry systems.

Back to List