Optimization of pixel size and propagation distance in X-ray phase-contrast virtual histology. (1st May 2022)

• Record Type: Journal Article
• Authors: Donato, S.; Arana Peña, L.M.; Bonazza, D.; Formoso, V.; Longo, R.; Tromba, G.; Brombal, L.
• Is Part Of: Journal of instrumentation. Volume 17:Number 5(2022)
• Publication Date: 2022-05-01
• Subjects: Computerized Tomography (CT) and Computed Radiography (CR) -- X-ray detectors -- Inspection with X-rays; Scientific apparatus and instruments -- Periodicals; 502.84
• DOI: 10.1088/1748-0221/17/05/C05021
• Languages: English
• ISSNs: 1748-0221

Abstract

Abstract: X-ray phase-contrast coupled to high-spatial resolution imaging systems provides a high sensitivity for distinguishing soft tissue structures in small samples, thus being suited for X-ray virtual histology. Propagation-based phase-contrast tomography can deliver a considerable gain in signal-to-noise ratio (SNR) at small pixel sizes when it is combined to a suitable phase retrieval filter. We optimized acquisition parameters, namely the propagation distance and the pixel size, with the aim of providing adequate spatial resolution and sensitivity for virtual histology of breast surgery specimens, scanned with a phase-contrast microtomography (μCT) system employing a commercial sCMOS detector at the SYRMEP beamline of the Italian synchrotron facility Elettra (Trieste, Italy). A pathological breast tissue sample was embedded in paraffin and imaged using a polychromatic synchrotron beam at an average energy of 24 keV. The high numerical optical aperture of the imaging system enabled to adjust the pixel size to 1, 2.5 and 4 μm. The scans were acquired at five sample-to-detector distances: 4.5, 150, 250, 500 and 1000 mm. SNR was measured in an homogeneous region portion of the μCT image for each combination of pixel size and propagation distance. Experimental results were compared to a theoretical model taking into account the actual point spread function of the employed imaging system. The measured gain of SNR associated with the application of the phase-retrieval matched the predictions for large Fresnel numbers ( N F > 2). For each pixel size, an optimal range of propagation distances was found. Optimal μCT reconstructions were then compared with their respective histopatological images, showing an excellent visibility of relevant structures. The optimization performed in this study will allow to select the most appropriate geometrical configurations for future acquisitions of virtual histology images of different specimens via phase-contrast microtomography.