International Joint Research Team Develops Technology to Reconstruct Internal Structures into 3D Images Using Just 3 X-rays Pictures
Presenting possibilities for application in medical diagnostics and industrial non-destructive testing
Published in Nature Communications
A joint Korea-US research team, including Professor Kim Kyeoung-hak of the Department of Chemical Engineering, has developed a technology capable of reconstructing internal structures into 3D images using only three X-ray photographs. The research team combined a thin and flexible 'paper-based X-ray detection film' with artificial intelligence (AI)-based 3D reconstruction technology. The related research results were published in the June issue of the international journal Nature Communications.
Existing X-ray detectors were made of rigid and flat substrates like glass or silicon, which posed a limitation as the images could be distorted when photographing curved body parts. Accordingly, the research team applied a method of directly growing lead-free perovskite microcrystals on the surface of cellulose fibers, a natural polymer, to produce a detection film that is thin, flexible, and can closely adhere to the curved surfaces of the human body.
By integrating AI technology here, they extracted structures of bones or regions of interest from 2D X-ray images taken from three different angles and reconstructed them into 3D structures. In fact, the research team reconstructed a bird's leg bone into a 3D image using just three X-ray photographs, and they also succeeded in reconstructing the position and arrangement of each pillar in a cylindrical structure containing multiple metal pillars.
This technology is characterized by its ability to obtain 3D information with a small number of images while utilizing existing standard X-ray imaging equipment. Consequently, it is expected to have potential applications in the future not only for medical diagnostics but also in the non-destructive testing fields of manufacturing industries such as aviation, automotive, and semiconductors.
Source : NewsH(https://www.newshyu.com/news/articleView.html?idxno=1025828)
Published in Nature Communications
A joint Korea-US research team, including Professor Kim Kyeoung-hak of the Department of Chemical Engineering, has developed a technology capable of reconstructing internal structures into 3D images using only three X-ray photographs. The research team combined a thin and flexible 'paper-based X-ray detection film' with artificial intelligence (AI)-based 3D reconstruction technology. The related research results were published in the June issue of the international journal Nature Communications.
Existing X-ray detectors were made of rigid and flat substrates like glass or silicon, which posed a limitation as the images could be distorted when photographing curved body parts. Accordingly, the research team applied a method of directly growing lead-free perovskite microcrystals on the surface of cellulose fibers, a natural polymer, to produce a detection film that is thin, flexible, and can closely adhere to the curved surfaces of the human body.
By integrating AI technology here, they extracted structures of bones or regions of interest from 2D X-ray images taken from three different angles and reconstructed them into 3D structures. In fact, the research team reconstructed a bird's leg bone into a 3D image using just three X-ray photographs, and they also succeeded in reconstructing the position and arrangement of each pillar in a cylindrical structure containing multiple metal pillars.
This technology is characterized by its ability to obtain 3D information with a small number of images while utilizing existing standard X-ray imaging equipment. Consequently, it is expected to have potential applications in the future not only for medical diagnostics but also in the non-destructive testing fields of manufacturing industries such as aviation, automotive, and semiconductors.
Source : NewsH(https://www.newshyu.com/news/articleView.html?idxno=1025828)