Head CT metal artifact correction method based on homomorphic adaptive learning

Abstract

The invention discloses a head CT metal artifact correction method based on homomorphic adaptive learning. The method comprises steps of constructing a deep learning generation framework by using a 17-layer convolutional neural network; performing feature extraction on the CT image by combining and using a VGG19 model; expanding the data set by using the prior information of CT continuous slices and the similarity of a CT structure; in order to solve the problem of medical data misalignment, improved anti-noise loss is used for the network, and the purposes of removing metal artifacts and retaining original disease information are achieved in the gradual iteration process by balancing the proportion of style loss and content loss; and result evaluation: training and testing are carried out on the clinical data set and the simulation data set, and the model is evaluated by comprehensively evaluating the method. According to the method, correction of the head CT artifacts is achieved, no new artifacts are generated and blurring is not caused while image detail information is reserved, the accuracy of clinical diagnosis is improved, and the method has good practical value.

Description

technical field [0001] The invention relates to the technical field of medical image processing, in particular to a head CT metal artifact correction method based on a deep learning convolutional neural network. Background technique [0002] In dental treatment, more and more patients choose metal implant surgery to improve oral problems. The computed tomography (CT) images of these patients are affected by high-density metals and produce artifacts, which may even cause misdiagnosis in severe cases, posing a great threat to subsequent treatment. [0003] Computed tomography (CT) is an advanced medical imaging technology that uses X-ray beams to scan specific areas of the human body to reconstruct lesions and provide important information for diagnosis. In dental treatment, it is often the case that a metal body is implanted in the tooth. However, since the metal body itself is a high-density substance, its existence will cause strong attenuation of X-rays during scanning im...

AI Technical Summary

Problems solved by technology

When the CT scan area contains multiple materials with different densities, the CT value cannot fully represent any one material, resulting in easily blurred areas at the interface between human tissue and metal implants, and streaking artifacts

Images