An objective image quality assessment method for optimizing medical image reconstruction parameters

Abstract

The invention discloses an objective image quality evaluation method for optimizing medical image reconstruction parameters. The present invention comprises the following steps: 1. adopting the method of loop topping to construct a plurality of virtual reference images, not only the quality of the reconstructed image can be analyzed by using the full reference image quality evaluation algorithm, but also parallel processing can be realized; 2. the combination of Daubechies wavelet transform and The eigenvalue decomposition analyzes the self-similarity of the image from different scales and different orientations; 3. The self-similarity of the reconstructed image obtained is used as the quality factor, and it is bubble-sorted to obtain the quality level of the reconstructed image, the highest quality level Corresponding to the optimal reconstruction parameters. The objective evaluation and subjective evaluation of the image quality proposed by the invention have good consistency, and especially can accelerate the optimization process of parameters in the medical image reconstruction algorithm.

Description

technical field [0001] The invention belongs to the technical field of medical image processing, in particular to an objective image quality evaluation method for optimizing medical image reconstruction parameters. Background technique [0002] Medical image quality evaluation has long been used by doctors to evaluate the performance of X-ray, computed tomography, magnetic resonance imaging, ultrasound imaging and many other modal imaging equipment. This evaluation is usually subjective and is called subjective image quality evaluation. In addition, a single imaging modality also has different reconstruction algorithms, and the quality of reconstruction also needs to be evaluated subjectively by doctors, which is time-consuming and laborious in the process of optimizing complex reconstruction parameters. In these application scenarios, objective image quality evaluation based on artificial intelligence and machine learning highlights its advantages due to its fast calculati...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the invention is to solve the quality level calibration problem of a series of reconstructed images with the same semantics

Provide an objective image quality evaluation method for optimizing medical image reconstruction parameters

The invention is based on the human visual system, making full use of the self-similarity of images with the same semantics, and extracting self-similar information. In order to solve the quality level calibration problem of a series of reconstructed images in the specific operation process, a loop top strategy is adopted combined with the quality factor The sorting method is used to obtain the quality level of a series of images, and a good image quality evaluation effect has been achieved.

Images