TwinTree Insert

14-04 Maximum-Intensity Projection

ith both time-of-flight and phase-contrast methods we now have isolated vas­­cu­­lar structures by acquiring individual two- or three-di­men­sio­nal ima­ges. However, an angiogram reflects entire vas­cu­lar structures in a two- or pseudo three-di­men­si­on­al overview.

In conventional x-ray and digital subtraction an­gio­gra­phy, this is created by sub­­trac­­tion of a mask from the contrast-enhanced an­gio­gra­phy images. In MR angiography, a special com­pu­ter al­go­rithm is used for this pur­po­se — the maximum-intensity pro­jec­tion (MIP).

Vascular structures on MR an­gio­gra­phy images show bright signal intensity.

The MIP algorithm allows the se­lec­tion of bright pixels in all parallel 2D slices or in the 3D slab or vo­l­ume and their projection into one image. The projection me­thod is similar to a shadowgram, with the exception that only high-signal-intensity pi­xels are projected into the final ima­ge. The corresponding pixels in each original image finally form the projection angiogram.

In 2D imaging, this is easily un­der­stan­dable as depicted in Fi­gu­re 14-15.

Figure 14-15:
The MIP operation. In this case, six slices have been acquired.

The figure shows how flow-related structures give the highest signal intensity (the big white circle on the final reconstructed image); however, there are other struc­tu­res visible with intermediate signal intensity (small gray circles on the ori­gi­nal ima­ges).

Only the highest valued pixel is represented on the final image. Thus, other struc­tu­res disappear.

In 3D imaging, this algorithm can be used to create images in any projection want­ed. On screen, one can visualize a rotating 3D angiogram.

spaceholder redMIP is a relatively simple and useful technique for processing angiographic MR data. It has some drawbacks such as the lack of discrimination between ar­te­ries and veins and high-signal-intensity non-vascular structures such as fat. New methods are being developed to overcome these problems, among them vessel tracking and volume rendering.

One disadvantage of MIP is that in routine clinical angiography bright non-vas­cu­lar tissues (e.g., fatty tissues) can represent highest signal intensity on the original pic­tures and thus be depicted on the angiogram. Such tissues can only be dis­cri­mi­na­ted from vascular structures by their anatomy.

spaceholder redFor black blood angiography, the contrary of MIP is done: those pixels with the lowest signal intensity are chosen, and with minimum-intensity projection (mIP) a black blood angiogram is created.

14-04-01 Reduction of Saturation Effects

Saturation effects are the gradual loss of T1 signal intensity by repeated ex­ci­ta­tion pulses. Too short TR leads to progressive loss of Mz; the same happens when the flip angle is increased.

Saturation effects associated with thick slabs can be partly overcome with a tech­ni­que called multiple overlapping thin slabs ac­qui­si­tion (MOTSA). Instead of one thick slab, several smaller slabs are acquired.

However, MOTSA creates the Venetian Blind artifact (Figure 14-16). Such artifacts do not exist when one uses Tilted Optimized Non-saturation Eex­ci­ta­tion, TONE. TONE applies ramped flip angles to the different slices of the slab. Increasing the flip angle counteracts saturation effects, in this case of slow flowing blood in deeper slices.

Figure 14-16:
MOTSA (multiple overlapping thin ac­qui­si­tion).

Combined with magnetization transfer contrast (MTC), which suppresses background sig­nal from brain parenchyma, TONE boosts the visibility of small vessels (Figure 14-17).

Figure 14-17:
Brain angiogram using TONE and MTC