The prior sections showed contrast filters for fast spin echo PD, T1, and T2 components separately. TR values are chosen to manipulate the T1 filter. TE values are chosen to manipulate the T2 filter. The total signal from a given tissue is the product of the relative signals (Y axis values) from each contrast filter. We discussed choosing TR and TE values to optimally distinguish different tissues from each other as well as to maximize sensitivity for detecting small changes from normal. These objectives required different strategies. This section looks at putting everything together for the fast spin echo sequence.
Start with the standard T1, T2, and PD weighted sequences used in neuro and MSK imaging.
A typical T1 weighted sequence in the brain has TR = 700ms and TE = 10ms.
The total signal from grey matter, on a scale from 0-1, is S(PD) x S(T1) x S(T2), or .
In this case, 0.8 x 0.39 x 0.9 = 0.28.
White matter signal is 0.7 x 0.56 x 0.87 = 0.39.
The grey matter is darker than the white matter on a "T1" weighted FSE with TR=700 and TE=10.
Comparing grey and white matter, the FSE sequence with TR=700ms and TE=10ms shows POSITIVE weighting for PD, i.e. increased PD from white matter to grey matter results in INCREASED signal, negative weighting for T1, i.e. increased T1 from white matter to grey matter results in DECREASED signal, and mininally positive weighting for T2; i.e. increased T2 from white to grey matter results in INCREASED signal.
The PD, T1, and T2 filters are not-synergistic for this FSE "T1" weighted sequence. The PD filter makes grey matter brighter than white matter. The T1 filter makes grey matter darker than white matter. The T2 filter makes grey matter brighter than white matter. It would be optimal, and make more sense, if the three filters could work together but we will discuss how to produce sequences that can do this later.
Grey matter is darker than white matter on this FSE sequence with TR=700ms and TE = 10ms because the negative T1 weighting is larger than the combined positive PD and T2 weighting. Since the T1 tissue filter dominates in the ranges of T1 and T2 seen in the brain, this is a T1 weighted image. The sequence is overall T1 weighted; the T1 filter predominates. We will present the math behind this in the next module.
How about a standard "T2" weighted sequence? Set TR=4000ms and TE=80ms.
Grey matter signal is 0.8 x 0.94 x 0.45 = 0.34.
White matter signal is 0.7 x 0.99 x .31 = 0.22.
Grey matter is brighter than white matter, but only mildy so.
In this sequence, the negative T1 weighting is overshadowed by the positive T2 and PD weighting resulting in grey matter being brighter than white matter. However, a closer look will show that for the T1 and T2 values of grey and white matter, there is also a lot of weighting from the proton density filter. We will show this mathematically in the next module. Changing T2 to the optimal value for distinguishing grey and white matter (94 ms as discussed previously) makes only a nominal difference. The slope of the curve between grey and white matter nearly as steep on the PD filter as it is on the T2 filter, especially when viewing on the logarithmic axes. This standard "T2" weighted image has a lot of PD weighting.
Lets now look at a standard "T1" weighted image for MSK. Reset TR to 700ms and TE to10ms. For this example it may be more useful to look at the graphs using the linear T1 axis.
For T1 and T2 values close to that of ligaments and tendons (T1 of 700 ms, T2 of 4 ms) the T1 filter is mildly negative (i.e. has a mild negative slope) and the T2 filter is steeply positive (steep positive slope). Small increases in T1 (from 700 ms to 750 ms) due to tendinopathy result in a mild decrease in signal due to T1. However, even small increases in T2 (from 4ms to 10 ms) result in large increases in signal due to the T2 filter. Increased PD due to edema/inflammation also produces increased signal.
The sequence with TR=700 ms and TE=10 ms is negatively T1 weighted in the tendons and muscles but very positively T2 weighted in the tendons! This is why conventional T1 weighted images are sensitive to tendinopathy: the T2 value of tendons is located along the steep portion of the T2 filter so small amounts of T2 prolongation in the tendons result in increased signal. This increased signal is not due to hemorrhage or protenaceous exudate in the tendon, it is due to edema. Calling it a "T1 weighted" image is probably a misnomer.
Note that a fluid filled tendon tear will NOT look bright on these images. Click on the fluid button. The sequence is still very negatively T1 weighted for fluid. The PD and T2 filters result in high signal but the strongly negative T1 filter brings that signal back down. A fluid filled tendon defect will be dark, as you would expect from a "T1" weighted image.
So lets now look at a "proton density" weighted FSE. Set TR=4000ms and TE=10ms and click on the ligament/tendon and simple fluid buttons. Pathology in a ligament, tendon, or meniscus results in small increases in PD, T1, and T2. A fluid filled tear replaces the structure with fluid signal resulting in large changes in PD, T1, and T2. The PD filter is positive for both processes. The T1 filter is flat for small increases in T1 and results in only moderately decreased signal for fluid. The T2 filter is markedly positive for both small and large increases in T2. Both tendinopathy (small increase in T2 and PD) and tear (large increase in T2 and PD) will be very bright. Therefore, distinguishing tendinopathy from a tear on PD images can be tricky.
We are taught that a rotator cuff tear in the shoulder should be as bright as fluid on both the PD and T2 weighted images. Why? On the "PD weighted" image fluid in the tear remains bright because the T1 filter still returns a signal of 0.65 and both large and small changes in T2 result in very high signal.
Change the TE back to 80ms for a "T2 weighted" image and see that for this curve small increases in T2 result in less signal gain than for large increases. On the this sequence tendinopathy is not as bright as a fluid because the T2 of tendinopathy (mildly elevated from normal) is much less than the T2 of fluid.
Input the parameters for some of your common sequences to see what are the relative weightings for different tissues.