Medical image parameter fitting method and device

Abstract

The invention belongs to the field of medical imaging and provides a medical image parameter fitting method and device so as to improve the stability of the parameter fitting effect during medical imaging. The method includes the steps that the range of a target image in a whole medical image is determined; fitting parameters of pixels are obtained during down-sampling of the target image; the target image is sampled in the mode that the resolution ratio is gradually increased, and the fitting parameters of the pixels during up-sampling of the target image are obtained till the resolution ratio reaches the original resolution ratio of the whole medical image. Compared with the prior art that the initial value of fitting parameters of pixels is selected subjectively, the technical scheme is the mode that the initial value of the fitting parameters of the pixels is self-adaptively adjusted, manual intervention is not needed, accordingly influences of subjective factors are eliminated, fitting results are more objective and stable, and the fitting effect is more ideal.

Description

technical field [0001] The invention belongs to the field of medical imaging, and in particular relates to a fitting method and device for medical image parameters. Background technique [0002] Quantifying relevant parameters of medical images by fitting models to collected data is a common means of processing medical images. In the process of model fitting, the selection of the initial value is very important, because an inappropriate initial value may cause the fitting to fall into a local minimum point, or the fitted parameters deviate from the true range, which will eventually lead to quantization errors. [0003] However, in the existing fitting methods of medical image parameters, the initial values ​​of relevant parameters are often selected based on the subjective feelings of the relevant R&D personnel. This selection method leads to strong subjectivity and poor stability of the results, and the fitting effect is not ideal. . Contents of the invention [0004] T...

AI Technical Summary

Problems solved by technology

In the process of model fitting, the selection of the initial value is very important, because an inappropriate initial value may cause the fitting to fall into a local minimum point, or the fitted parameters deviate from the true range, eventually leading to quantization errors

[0003] However, in the existing fitting methods of medical image parameters, the initial values ​​of relevant parameters are often selected based on the subjective feelings of the relevant R&D personnel. This selection method leads to strong subjectivity and poor stability of the results, and the fitting effect is not ideal.

Images