Source: Suzhou Institute of Biomedical Engineering, Chinese Academy of Sciences
At present, with the increasing application of CT tomography in medical diagnosis more widely, CT radiation will cause some damage to the human body, and increase the incidence rate of cancer and other diseases, so we need to ensure the image quality under the premise of as far as possible to reduce the radiation dose of X radiation. In order to reduce the radiation dose, Naidich et al first proposed the concept of low dose CT in 1990, in the same other scanning parameters, by reducing the tube current imaging in diagnosis and to reduce the radiation dose required. When the CT tube current is too low, the number of photons received by the detector will be reduced, the reconstruction image will have serious noise interference, produce streak artifact (streak artifacts), so we need to suppress the noise in the image, including noise filtering on image reconstruction and denoising and improving the reconstruction algorithm three a method of projection data. The issue of low-dose imaging in CT imaging has attracted more and more attention. A large number of clinical practice has shown that over normal range of CT radiation dose is closely related to abnormal metabolism of the human body, and even cancer and other diseases. However, in the current CT devices, the images acquired by low dose scan protocols can cause serious noise and artifact of the reconstructed images, and affect the clinician's correct rate of abnormal tissue. The low dose imaging protocol mainly includes: 1) using reduced sampling exposure; 2) using lower tube pressure and tube current exposure. The former is more suitable for open CBCT devices, while the latter is more suitable for helical CT devices. Therefore, it is of great academic and clinical value to reconstruct the high quality images by using the original projection data collected by low dose protocols.
Li Ming et al X ray image group Medical Research Institute for biomedical engineering of Suzhou Academy of Sciences Chinese chamber put forward the regularization function model based on SL0, the model is based on the noise properties of the reconstructed image, adaptive regularization constrains, which has better noise suppression and protection characteristics for the soft tissue. The method can achieve ideal reconstruction results in simulation experiment, clinical data experiment and animal data reconstruction.
Relevant research results are published in international biomedical research (BioMed, Research, International, Volume, 2016, Article, ID 2180457, 12, pages). The achievement has been supported by national key research and development plan, digital diagnosis and treatment project (approval number: 2016YFC01035022016YFC0104505) and Natural Science Foundation of Jiangsu (BK20151232).