Learning Perspective Deformation in X-Ray Transmission Imaging
This addresses geometric accuracy issues in medical imaging for clinicians, but it is incremental as it builds on existing networks like Pix2pixGAN and TransU-Net.
The paper tackled perspective deformation in cone-beam X-ray imaging by using two complementary views to correct geometric distortions, achieving better performance with Pix2pixGAN in polar space and TransU-Net in Cartesian space, and demonstrating robustness on synthetic and real data including cases with metal implants.
In cone-beam X-ray transmission imaging, perspective deformation causes difficulty in direct, accurate geometric assessments of anatomical structures. In this work, the perspective deformation correction problem is formulated and addressed in a framework using two complementary (180°) views. The complementary view setting provides a practical way to identify perspectively deformed structures by assessing the deviation between the two views. It also provides bounding information and reduces uncertainty for learning perspective deformation. Two representative networks Pix2pixGAN and TransU-Net for correcting perspective deformation are investigated. Experiments on numerical bead phantom data demonstrate the advantage of complementary views over orthogonal views or a single view. They show that Pix2pixGAN as a fully convolutional network achieves better performance in polar space than Cartesian space, while TransU-Net as a transformer-based hybrid network achieves comparable performance in Cartesian space to polar space. Further study demonstrates that the trained model has certain tolerance to geometric inaccuracy within calibration accuracy. The efficacy of the proposed framework on synthetic projection images from patients' chest and head data as well as real cadaver CBCT projection data and its robustness in the presence of bulky metal implants and surgical screws indicate the promising aspects of future real applications.