Low-dose CT image reconstruction based on a deep neural network

Computed tomography (CT) technology plays an important role in the field of medicine. It helps the doctor give the accurately judgement of lesions through the clear CT image. But the patients can't afford a large amount of radiation to obtain the clear CT image. Therefore, the problem of low-dose CT image reconstruction is particularly prominent. This paper presents a low-dose CT image reconstruction method based on deep learning algorithm. The multi-layers convolution network is constructed to deeply mine the features of the input data, so as to weaken the noise signal in the original data and obtain a clear and identifiable reconstructed image. This method aims at the processing of low-dose projection data, avoids the radiation damage of the irradiated party, and reduces the energy loss of the equipment. Through the analysis of the experimental results, this paper summarizes the future work in image detail information capture and database expansion, so as to improve the overall image reconstruction quality.