Patient-specific reconstruction of volumetric computed tomography images from a single projection view via deep learning

• Published in: Nature biomedical engineering Vol. 3; no. 11; pp. 880 - 888
• Main Authors: Shen, Liyue, Zhao, Wei, Xing, Lei
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 01.11.2019; Nature Publishing Group
• Subjects: 59; 639/166/985; 692/4028/67/1059/485; 692/700/1421/1846/2771; 692/700/1421/2025; Abdomen - diagnostic imaging; Anatomy; Angular position; Biomedical and Life Sciences; Biomedical Engineering/Biotechnology; Biomedicine; Biopsy; Biopsy, Needle; Computed tomography; Cone-Beam Computed Tomography - methods; Deep Learning; Head - diagnostic imaging; Humans; Image reconstruction; Imaging, Three-Dimensional - methods; Lung - diagnostic imaging; Machine learning; Medical imaging; Neck - diagnostic imaging; Projection; Radiation therapy; Radiographs; Radiography; Radiotherapy; Radiotherapy Planning, Computer-Assisted - methods; Tomography; Two dimensional models
• ISSN: 2157-846X; 2157-846X
• DOI: 10.1038/s41551-019-0466-4

Tomographic imaging using penetrating waves generates cross-sectional views of the internal anatomy of a living subject. For artefact-free volumetric imaging, projection views from a large number of angular positions are required. Here we show that a deep-learning model trained to map projection radiographs of a patient to the corresponding 3D anatomy can subsequently generate volumetric tomographic X-ray images of the patient from a single projection view. We demonstrate the feasibility of the approach with upper-abdomen, lung, and head-and-neck computed tomography scans from three patients. Volumetric reconstruction via deep learning could be useful in image-guided interventional procedures such as radiation therapy and needle biopsy, and might help simplify the hardware of tomographic imaging systems. A deep-learning model trained to map 2D projection views of a patient to the corresponding 3D anatomy can subsequently generate volumetric tomographic X-ray images of the patient from a single projection view.