The dual-gain mode: a way to enhance the dynamic range of X-ray detectors
Evangelos Matsinos, Wolfgang Kaissl
Abstract
Varian Medical Systems has manufactured and recently put into operation a clinically-applicable solution for image-guided radiation therapy. Cone-beam CT imaging, one of the operation modes of the imaging unit of this device, aims at high-quality volumetric reconstruction. To boost the image quality, the dual-gain mode, a successful means for enhancing the dynamic range of the flat-panel detectors and obtaining better results in the contrast of the reconstructed image, was developed and successfully tested during the last few years. The important steps in the calibration of this mode involve a correction to the pulse widths associated with the X-ray production, the assessment of the detector signal above which nonlinear effects become significant and the determination of some properties of the detector pixels, namely, of dark fields, flatness corrections, etc. Finally, the defect-pixel map is obtained, containing dead and flickering pixels, as well as pixels with properties which are sufficiently `out of range'. An effect observed in the offset correction of raw images seems to originate in the mismatch between the measured and extrapolated values of the dark field; a correction scheme is proposed to take account of this effect. In the data processing, which is achieved on the basis of the files produced in the calibration of the mode, the high-gain signal is used whenever meaningful; otherwise, it is substituted by the low-gain signal, properly scaled. Defect pixels are interpolated from their good neighbours; a procedure achieving a correction in case of small defect-pixel clusters is outlined.
Create a lesson
Related papers
Unimodality and Radial Monotonicity of the Magnetic Resonance Fingerprinting T1/T2 Matching Objective
Ze Wang
A Reconfigurable Pipelined-SAR ADC with Embedded Compression for Temporal Compressed-Sensing Ultrasound Imaging
Reza Pakdaman Zangabad, Xitie Zhang, Levent Degertekin et al.
Physics-Assisted Deep Learning Denoising for Stabilized IMPULSED dMRI Microenvironment Parameter Fitting
Wen Li, Yan Dai, Arely Perez Rodriguez et al.
Constrained estimation of rotational invariants of the cumulant expansion (RICE) for rapid tensor-valued diffusion MRI
Jinyang Yu, Oliver Gödicke, Frederik B. Laun et al.
Dose-PlanNet: Physics Based Radiotherapy Dose Prediction with Deep Learning
Ankit Bhattacharjee, Sougata Maity, Santam Chakraborty et al.
Sparse Delta Integration method for the calculation of spatiotemporal pressure fields of arbitrary ultrasound transducer geometries
Deyver E. Rivera, Charlie Demene, Mickael Tanter