Generation of Pseudo X-Ray and Digitally Reconstructed Radiograph from CT images

Computed Tomography (CT) scanning offers 3D insights into the human body but involves higher radiation exposure than traditional X-Rays, increasing the relevance of 3D reconstructions from 2D data. In this work, datasets of CT- pseudo-X-Ray and CT- Digitally Reconstructed Radiograph (DRR) pairs are generated using ray-sum tracing and projection techniques. This can be helpful for implementing machine learning and deep learning architecture as 2-D data availability is limited. The described approach generates CT-pseudo-X-Ray and CT-Digitally Reconstructed Radiograph (DRR) pairs using ray-sum tracing and projection methods. The pseudo-X-Ray is created by combining bone and soft tissue projections, while the DRR is derived via ray-sum tracing. By retaining the CT images in Digital Imaging and Communications in Medicine (DICOM) format, the created dataset enhances robustness and efficiency, supporting 3D reconstruction-a rapidly advancing field.