Abstract:Objective. X-ray diffraction (XRD) has been considered as a valuable diagnostic technology providing material specific ‘finger-print’ information i.e. XRD pattern to distinguish different biological tissues. XRD tomography (XRDT) further obtains spatial-resolved XRD pattern distribution, which has become a frontier biological sample inspection method. Currently, XRD computed tomography (XRD-CT) featured by the conventional CT scan mode with rotation has the best spatial resolution among various XRDT methods, but its scan process takes hours. Meanwhile, snapshot XRDT methods such as coded-aperture XRDT (CA-XRDT) aim at direct imaging without scan movements. With compressed-sensing acquisition applied, CA-XRDT significantly shortens data acquisition time. However, the snapshot acquisition results in a significant drop in spatial resolution. Hence, we need an advanced XRDT method that significantly accelerates XRD-CT acquisition and still maintains an acceptable imaging accuracy for biological sample inspection. Approach. Inspired by the high spatial resolution of XRD-CT from rotational scan and the fast compressed-sensing acquisition in snapshot CA-XRDT (SnapshotCA-XRDT), we proposed a new XRDT imaging method: sparse-view rotational CA-XRDT (RotationCA-XRDT). It takes SnapshotCA-XRDT as a preliminary depth-resolved XRDT method, and combines rotational scan to significantly improve the spatial resolution. A model-based iterative reconstruction (MBIR) method is adopted for RotationCA-XRDT. Moreover, we suggest a refined system model calculation for the RotationCA-XRDT MBIR which is a key factor to improve reconstruction image quality. Main results. We conducted our experimental validation based on Monte-Carlo simulation for a breast sample. The results show that the proposed RotationCA-XRDT method succeeded in producing good images for detecting 2 mm square carcinoma with a 15-view scan. The spatial resolution is significantly improved from current SnapshotCA-XRDT methods. With our refined system model, MBIR can obtain high quality images with little artifacts. Significance. In this work, we proposed a new high spatial resolution XRDT method combining coded-aperture compressed-sensing acquisition and sparse-view scan. The proposed RotationCA-XRDT method obtained significantly better image resolution than current SnapshotCA-XRDT methods in the field. It is of great potential for biological sample XRDT inspection. The proposed RotationCA-XRDT is the fastest millimetre-resolution XRDT method in the field which reduces the scan time from hours to minutes.

Optimization of the Structured Illumination Series for Compressive X-Ray Tomosynthesis

Non-Linear 3d Reconstruction For Compressive X-Ray Tomosynthesis

Instant Coded X-ray Computed Tomography Via Nonlinear Reconstruction

K-edge Coded Aperture Optimization for Uniform Illumination in Compressive Spectral X-ray Tomosynthesis

A Wavelet-Based Single-View Reconstruction Approach for Cone Beam X-Ray Luminescence Tomography Imaging.

Coded Aperture Optimization in X-Ray Tomography via Sparse Principal Component Analysis

Image reconstruction and scan configurations enabled by optimization-based algorithms in multispectral CT

Nonconvex Compressive Sensing for X-Ray Ct: an Algorithm Comparison

Optimization of Image Quality and Acquisition Time for Lab-Based X-ray Microtomography Using an Iterative Reconstruction Algorithm

Algorithm-driven Advances for Scientific CT Instruments: From Model-based to Deep Learning-based Approaches

Integer Optimization of CT Trajectories using a Discrete Data Completeness Formulation

Optimization-based image reconstruction with artifact reduction in C-arm CBCT.

Optimization-Based Image Reconstruction from Low-Dose Patient Breast Ct Data

Tomosynthesis reconstruction from multi-beam X-ray sources

Optimization methods for very accurate Digital Breast Tomosynthesis image reconstruction

CT Reconstruction from Simultaneous Projections: A Step towards Capturing CT in One Go

A compressed sensing approach to low-radiation CT reconstruction

Compressive Spectral X-Ray CT Reconstruction Via Deep Learning.

Method of Sparse-View Coded-Aperture X-Ray Diffraction Tomography

Structural Prior Enhanced Compressed Sensing For Ct Reconstruction With Incomplete Data

Preliminary Investigation of Optimization-Based Image Reconstruction for TOF PET with Sparse Configurations