Basic process of medical image registration

The basic process of medical image registration is as follows:

Firstly, feature extraction is carried out on two images to obtain feature points; Matching feature point pairs are found by similarity measurement; Then, the image space coordinate transformation parameters are obtained through the matched feature point pairs; Finally, the coordinate transformation parameters are used for image registration.

basic concept

When doing medical image analysis, it is often necessary to analyze multiple images of the same patient together in order to obtain comprehensive information of patients in many aspects and improve the level of medical diagnosis and treatment. To quantitatively analyze several different images, we must first solve the problem of strict alignment of these images, which is what we call image registration.

Medical image registration refers to seeking a kind of (or a series of) spatial transformation for one medical image to make it consistent with the corresponding points on another medical image in space. ?

This consistency means that the same anatomical point on the human body has the same spatial position in two matched images. The result of registration should make all anatomical points on the two images, or at least all points with diagnostic significance, match the points of surgical interest.

Medical image registration technology is an important branch of medical image processing developed in 1990s. Medical image registration technology mainly discusses post-acquisition registration after data acquisition, also known as retrospective registration? .

At present, the international research on medical image registration mainly focuses on tomographic images, such as CT and MRI. Nuclear magnetic resonance, single photon emission tomography, positron emission tomography, etc. ) and time series images (such as fMRI and 4D cardiac images).