120 121 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 2 0 / 2 1 D I A M O N D L I G H T S O U R C E A N N U A L R E V I E W 2 0 2 0 / 2 1 Lowdose X-ray single-pixel imaging at high frame rates Related publication: Sefi O., KleinY., Strizhevsky E., Dolbnya I. P. & Shwartz S. X-ray imaging of fast dynamics with single-pixel detector. Opt. Express 28 , 24568 (2020). DOI: 10.1364/oe.396497 Publication keywords: X-ray; Single-pixel imaging; Ghost imaging; Lens less X -ray imaging methods have numerous applications in various areas, ranging from basic science to medicine and security. The main advantage of using X-rays for imaging is their unique capability to penetrate through surfaces. Most X-ray pixelated detectors are used for static applications. However, there is agrowingneed for detectors that combinehigh spatial resolutionwithhigh frame rates.These detectorswill allowthe application of X-rays formeasurements of dynamic systems, including acousticwaves inmatter, phase transitions and medical imaging. In traditional imaging methods that use pixelated detectors, the readout time increases with the number of pixels. This means there is a fundamental trade-off between the frame rate and the number of pixels. Single-pixel detectors can be much faster than pixelated detectors but cannot provide spatial information directly. Researchers from Bar-Ilan University in Israel investigated a novel imaging technique that offers high frame rate imaging at high resolution based on thermal (or pseudothermal) ghost imaging (GI). As the concept had never been demonstrated with X-rays, they used B16 - a beamline at Diamond Light Source dedicated to testing new techniques and concepts. They successfully demonstrated the imaging of a rotating object at a frame rate of 100 kHz and a resolution of about 40 μm. This technique can be applied for medical imaging where measurements of fast dynamics are required, e.g. non-invasive cardiac imaging, and for non-destructive imaging of movingmechanical components. X-ray imaging methods are used for numerous applications in a variety of areas. Recently, there is a growing need for detection schemes with combined high spatial resolution and high frame rates, which are essential for the understanding of numerous phenomena and processes and can lead to novel applications. However, there is a fundamental trade-off between the resolution and the frame rate, which rectrices the capabilities of two-dimensional (2D) X-ray detectors. A novel imaging technique, which is based on the method of thermal ghost imaging (GI) offers high frame rate imaging at high resolution. GI is a single-pixel imaging technique, which uses intensity fluctuations, which are introduced into the beam by a diffuser. In the current demonstration of this method, which resembles the computational GI approach 1 , the measurements are performed in two steps: first, the distributions of the intensity fluctuations for the various realisations are recorded by a 2D detector in the absence of the object. In the second step, the object is inserted, and the measurements of the test beam are recorded by a single-pixel detector. After the two sets of measurements are completed, the image is reconstructed by correlating the measurements from each realisation. In the present work we extended the applicability of GI for periodically moving objects 2 by simply measuring a sequence of tests at high sampling frequency for each position of the diffuser. Each test sequence is taken along the entire period of the motion of the object. Thus, each position of the diffuser produces a set of test signals, corresponding to different positions along the trajectory of the object. Finally, each frame, which corresponds to a specific position of the object, is reconstructed separately. This work is the first experimental demonstration of X-ray GI for fast dynamics, where the motion of a chopper spinning at 200 Hz was captured with a frame rate of 100 kHz, a spatial resolution of about 40 μm, and a field of view of 0.6 mm× 0.6 mm. The experiment was conducted on B16 with a monochromatic beam at 9 keV. The experimental setup consists of a sandpaper diffuser, a rotating optical chopper, a slow 2D detector for the reference measurements and a single-pixel detector for the test measurements. For the reconstruction a compressive sensing algorithm 3,4 was used. A schematic description of the experimental setup is shown in Fig. 1. Several snapshots from the reconstructed movie of the rotating chopper are shown in Fig. 2. The number of pixels in each frame is 8550 and the frame rate is 10 5 frames per second. The reconstruction was done using 4900 realisations per each frame with an average of 8.5·10 3 counts per realisation. In the upper left frame, the blade of the chopper blocks almost the entire beam except from a small area near the bottom right corner.The next panel shows the 6 th later frame where the chopper blocks a smaller portion of the beam. The rest of the frames show the motion of the chopper at measurement times that are indicated in Fig. 2 until the bottom right panel. One important aspect of GI is the possibility to reconstruct the object under lowdoses of radiation. In Fig. 3 the dependence of the signal-to-noise ratio (SNR) on the number of the photons detected by the test detector is shown. Here, the blue dots indicate the SNR calculated from the measured data and the error bars indicate the counting statistics. The insets (a)-(c) are the corresponding reconstructed images for different radiation doses. It is clear that satisfactory reconstructions were achieved under low dose conditions. To summarise, a demonstration of the ability to use the method of GI for high-resolution large field of view X-ray imaging of fast dynamics was made, as well as the possibility for low dose imaging, and therefore can be implemented with conventional X-ray tubes. References: 1. KleinY. et al. X-ray computational ghost imaging with single-pixel detector. Opt. Express 27 , 3284 (2019). DOI: 10.1364/oe.27.003284 2. ZhaoW. et al. Ultrahigh-Speed Color Imaging with Single-Pixel Detectors at Low Light Level. Phys. Rev. Appl. 12 , 34049 (2019). DOI: 10.1103/ PhysRevApplied.12.034049 3. Katz O. et al. Compressive ghost imaging. Appl. Phys. Lett. 95 , 131110 (2009). DOI: 10.1063/1.3238296 4. Li C. et al. An efficient augmented Lagrangianmethod with applications to total variation minimization. Comput. Optim. Appl. 56 , 507–530 (2013). DOI: 10.1007/ s10589-013-9576-1 Funding acknowledgement: EU Framework Program for Research and Innovation Horizon 2020 Framework Programme (CALIPSOplus 730872). Corresponding authors: Prof. Sharon Shwartz, Bar-Ilan University, sharon.shwartz@biu. ac.il; Dr Or Sefi, Bar Ilan University,
[email protected] Optics andMetrology Group Beamline B16 Figure 1: A schematic description of the experimental system for the (a) reference and (b) test measurements. Figure 2: Reconstructed movie frames of an optical chopper rotating at 200 Hz. The movie is recorded at a frame rate of 10 5 frames per second at resolution of about 40 µm. Figure 3: Dependence of the signal-to-noise ratio on the flux of the test beam for one frame from the movie. The horizontal error bars indicate the counting statistics. In the insets (a-c) the reconstructed images corresponding to 30, 375 and 1160 counts at the detector, are shown respectively.