XAP-Lab: A software tool for designing flexible X-ray acquisition protocols

Abstract

Background and objectiveThe availability of digital X-ray detectors, together with the development of new robotized hardware and reconstruction algorithms, opens the opportunity to provide 3D capabilities with conventional radiology systems. This would be based on the acquisition of a limited number of projections with non-standard geometrical configurations. The versatility of these techniques is enormous, enabling the introduction of tomography in situations where a CT system is hardly available, such as during surgery or in an ICU, or in which a reduction of radiation dose is key, as in pediatrics. Computer simulations are a valuable tool to explore these possibilities before their actual implementation on real systems. Existing software tools generally simulate only standard acquisition protocols, such as cone-beam with circular trajectory, thus not allowing the users to evaluate more sophisticated projection geometries. The goal of this work is to design a simulation tool that enables the design of acquisition protocols with flexible projection geometries. MethodsWe present XAP-Lab, a software tool for the design of X-ray acquisition protocols with flexible trajectories. For a given projection geometry, defined through a graphical user interface, it allows the user to simulate projections using GPU-accelerated kernels, the visualization of the scanned field of view and the estimation of the total radiation dose. The complete acquisition protocol can then be exported with the appropriate format for its use on real systems.We tested the software by optimizing a tomosynthesis protocol and validating the results with real acquisitions using a SEDECAL NOVA FA radiography system and phantoms for quantitative and qualitative evaluation. ResultsQuantitative evaluation using a phantom showed a mean error under 4 mm for each position, below the ±5 mm tolerance of the system specified by the manufacturer. Visual evaluation on a thorax acquisition also showed a good geometrical agreement between simulated and real projections. ConclusionsResults showed an excellent matching with simulations, supporting the usefulness of XAP-Lab for the design of new acquisition protocols with non-standard geometries.