CT Film Recovery via Disentangling Geometric Deformation and Illumination Variation: Simulated Datasets and Deep Models
This work tackles the practical problem of recovering usable medical images from photographs of CT films, which is relevant for patients and clinicians in regions where film prints are common for storage and consultation.
This paper addresses the problem of recovering CT film images that suffer from geometric deformation and illumination variation. The authors created a large-scale simulated dataset, CTFilm20K, and proposed a deep learning framework to disentangle and correct these distortions, outperforming previous approaches on both simulated and real images.
While medical images such as computed tomography (CT) are stored in DICOM format in hospital PACS, it is still quite routine in many countries to print a film as a transferable medium for the purposes of self-storage and secondary consultation. Also, with the ubiquitousness of mobile phone cameras, it is quite common to take pictures of the CT films, which unfortunately suffer from geometric deformation and illumination variation. In this work, we study the problem of recovering a CT film, which marks the first attempt in the literature, to the best of our knowledge. We start with building a large-scale head CT film database CTFilm20K, consisting of approximately 20,000 pictures, using the widely used computer graphics software Blender. We also record all accompanying information related to the geometric deformation (such as 3D coordinate, depth, normal, and UV maps) and illumination variation (such as albedo map). Then we propose a deep framework to disentangle geometric deformation and illumination variation using the multiple maps extracted from the CT films to collaboratively guide the recovery process. Extensive experiments on simulated and real images demonstrate the superiority of our approach over the previous approaches. We plan to open source the simulated images and deep models for promoting the research on CT film recovery (https://anonymous.4open.science/r/e6b1f6e3-9b36-423f-a225-55b7d0b55523/).