Multi-modal medical image auxiliary diagnosis method

Abstract

The invention discloses a multi-modal medical image auxiliary diagnosis method. The method comprises the following steps: adopting a convolutional neural network to realize identification and pre-judgment of a focus part; then, improving the image quality through an image enhancement technology, establishing different coordinate systems through an electromagnetic positioning system, achieving theregistration of medical images of different modes in combination with an ICP registration algorithm, and finally achieving the multi-mode medical image fusion. By means of the method, interference ofhuman subjective factors can be avoided, the advantages of different image technologies can be integrated, recognition and judgment of focus parts are improved, and therefore medical diagnosis efficiency and accuracy are improved; moreover, the method can be applied to imaging equipment with multiple modes, such as CT, ultrasound, MRI, PECT and the like.

Description

technical field [0001] The present invention relates to the technical field of multimodal medical images, in particular to a multimodal medical image aided diagnosis method. Background technique [0002] With the development of technology, the subject of medical imaging is becoming more and more important in clinical application. Since image fusion technology is still a brand new field, there are many technologies that have not yet been conquered. Image registration is currently recognized as a difficult image processing technology, and it is also a technical difficulty that restricts the development of medical image fusion technology and the use of medical image fusion software. However, image fusion can perform medical images from multiple different modalities. Organic fusion, making full use of the complementarity and redundancy of different types of medical images to describe lesions, achieving the effect of 1+1>2, better assisting doctors to understand the comprehen...

AI Technical Summary

Problems solved by technology

[0004] 1) Various medical imaging devices in hospitals have different visualization effects on different organs due to their different imaging principles. For example, CT images have particularly good resolution and resolution for tissues with obvious differences in calcium content, but for muscle tissue The resolution and resolution of ultrasound are not as good as that of ultrasound. Most of the various medical imaging equipment are used alone, and the image effect formed by using them alone has great limitations;

[0005] 2) The shortcomings brought about by single-mode medical imaging equipment sometimes have a greater negative impact on the diagnostic process, such as: the outline of the lesion is incomplete or has false shadows, and the amount of information is insufficient;

[0006] 3) Disease prediction through medical images is mainly through manual analysis, which is easily affected by subjective factors and brings greater uncertainty to subsequent registration and image fusion processing

Images