Content: chemical shift artifact: refers to the image artifact caused by chemical shift. Due to the existence of chemical shift, the precession frequency of hydrogen protons in water molecules is about 3.5ppm higher than that in fat, equivalent to 150Hz / T. The precession frequency difference of 1.5T equipment is about 225hz, and that of 1.0T equipment is about 150Hz.[knowledge foreshadowing] the spatial coding of image is realized by gradient field. Our machine usually has a set of three sets of gradient fields. These three groups of gradient fields are actually three coils, which are located in sagittal, axial and coronal positions. Why are they three groups? What do they have to do with chemical? Because there are three groups, it is enough for us to carry out spatial positioning: suppose we want to form a transverse image, we first select the layer with GZ gradient field. Through the synergy of gradient field and main magnetic field, we choose one layer. Why can we choose one floor? We superimpose the gradient field and the main magnetic field to make the magnetic field intensity of each point on the main magnetic field different. We know that the resonance condition is the same frequency. If we choose a point where the sum of the magnetic field strength (the sum of the main magnetic field and the gradient magnetic field vector) is 42.58 MHz, there is only one proton energy level on the long axis of the human body that is the same as 42.58 MHz and can resonate. It can absorb the resonance of the RF pulse, then relax and transmit the signal, the coil induced signal, and finally the image; The proton precession frequency of other energy levels is different from 42.58 MHz, and resonance will not occur. How can we imagine no resonance? In this way, we choose a layer.
As shown in the figure, the horizontal solid line is the main magnetic polenetic field of 1.0T. If the magnetic field is an absolutely uniform ideal state, we know that every point on this line is continuous and the value is 1.0T; The dotted line is the gradient magnetic field. The left side is low and the right side is high. This low and high do not represent the size, but only the direction. The low side is opposite to the direction of the main magnetic field and plays the role of weakening the main magnetic field. Then the main magnetic field on the left side of the intersection point is weakened by the gradient field, and the magnetic field intensity on the left side is lower. Similarly, the gradient field on the right side of the intersection point has the same direction as the main magnetic field, Then the main magnetic field on the right plays the role of increasing the main magnetic field. The more to the right, the greater the strength of the main magnetic field. This is what we need to know first. The gradient field at the intersection point is 1t after the superposition of 0,0 and 1t of the main magnetic field, which means that the main magnetic field is not affected by the gradient magnetic field at this point. After understanding the superposition of gradient field and main magnetic field, we understand that as long as there is gradient field, the magnetic field intensity after their action is not a fixed value, but a magnetic field change value that changes linearly from left to right and from low to high. From the above figure, we know that the change of magnetic field is linear, the lowest on the left, the highest on the right and the linear change in the middle. Assuming that the human body is examined in the magnetic field, the resonance frequency of hydrogen proton is 42.58mhz under 1t magnetic field intensity, then there can only be one point in the middle of the linear magnetic field is 42.58mhz, the left is lower than 42.58mhz and the right is higher than 42.58mhz, Therefore, only one layer can resonate, and other layers will not resonate. Resonance generates signals, and non resonance has no signals. This is the method. We have completed layer selection, but only layer selection. The positions of each pixel point (voxel point) on this layer are chaotic, there is no order, but the precession frequency is the same. Now we can’t image, We have two more steps: phase coding and frequency coding.
Phase coding: before frequency coding, phase coding should be carried out first. By superimposing gradient field, all protons in the system that meet the resonance conditions resonate and rotate and swing. However, due to different precession frequencies (caused by different magnetic field strength), some protons precession fast and some precession slow. Precession is rotation and swing, Their phase will also be different due to the difference of precession frequency, that is, the direction of vector is different (as shown in the figure, the arrow direction is the phase direction),
At this time, remove the gradient field. After removing the gradient field, the magnetic field is not superimposed, but the strength of a single main magnetic field, and the strength of the main magnetic field is uniform and the same. At this time, all proton precession frequencies change from different to the same. Although the precession frequency is the same, different phases are saved, that is, the precession frequency is the same, but the direction of rotation and swing is different, which completes the phase coding.
After the phase coding is completed, the frequency coding starts. The frequency coding needs to add gradient field in the vertical direction. As can be seen from the above figure, after the phase coding, each vertical row has the same phase and the adjacent two rows have different phases, so the frequency coding needs to add gradient field in the vertical direction, The gradient field results in different precession frequencies of pixel protons in the same phase (as shown in the figure below),
This uses the same principle. Different precession frequencies lead to different phases and frequencies of each pixel, so as to complete the coding of each pixel. If you want the computer to correctly reconstruct the image, you need to have a difference between each pixel before you can put it where it should be. If the phase and frequency of each pixel are the same, there is no difference. The computer cannot reconstruct the image. Only if each pixel is unique, the computer can make a puzzle. Therefore, we know the coding order. In the first step, the gradient field selects the layer, the second gradient field carries out phase coding, and the third gradient field carries out frequency coding. Only by understanding the principle of coding can we better understand the chemical shift artifact.