Spin Echo vs Gradient Echo
- Spin echo uses a 180° "refocusing" pulse to undo fixed field imperfections, so it ignores them. Gradient echo doesn't, so it sees them.
- Because gradient echo skips that 180° pulse, it can flip and read out the signal much faster — speed is its whole selling point.
- The flip side: gradient echo stays sensitive to anything that messes with the local field (metal, air, blood products, calcium), which makes it the go-to for hunting hemorrhage and susceptibility.
- Spin echo is the slower, cleaner, more "honest" workhorse; gradient echo is the fast, twitchy specialist.
If you've ever been confused about why one MRI sequence takes four minutes and another takes twenty seconds, why one makes a tiny metal clip look like a black cannonball and the other barely notices it — congratulations, you've bumped into the spin echo versus gradient echo divide. It's one of those forks in MRI physics where a single design choice ripples out into everything you see on the image. Let me walk you through it.
The problem both sequences are trying to solve
After you tip the protons over with a radiofrequency pulse, they spin together in a neat little chorus, all in phase. The signal we read out depends on them staying in sync.
They don't stay in sync. They fan out and lose phase with each other — this is what makes the signal decay. Some of that fanning is genuine, irreversible tissue property (true T2 relaxation). But a big chunk is cheating: it comes from the magnet's field being slightly uneven from spot to spot. A proton in a stronger pocket spins a hair faster; its neighbor in a weaker pocket lags. They drift apart not because the tissue is interesting, but because the magnet is imperfect.
That recoverable, magnet's-fault dephasing is the whole battleground. Spin echo erases it. Gradient echo lives with it.
Spin echo: the magic of the U-turn
Here's the trick that makes spin echo so clean. Picture a foot race where some runners are slightly faster and some slightly slower, so they spread out down the track. Now, at a precise moment, you blow a whistle and everyone turns around and runs back at the same pace they were going.
What happens? The fast runners are now behind — they have the farthest to go. The slow runners are ahead but pokey. And they all cross the original start line at the exact same instant. The spread undoes itself.
That whistle is the 180° refocusing pulse. By flipping the spins halfway through, spin echo makes the fast and slow protons swap their head starts, so all the magnet-imperfection dephasing cancels out and the signal "rephases" into an echo. Because those fixed field flaws get undone, the contrast you're left with reflects true T2 — the honest tissue property.
This is why spin echo is the reliable workhorse for routine anatomy. It shrugs off static field imperfections and metal artifact better than gradient echo. The cost is time: that 180° pulse and the rephasing take a while, and stacking many of them is slow.
Gradient echo: skip the U-turn, gain speed
Gradient echo asks an impatient question: what if we don't bother with that 180° pulse?
Instead of a refocusing whistle, gradient echo uses the magnetic field gradients themselves to deliberately dephase the spins and then reverse the gradient to rephase them into an echo. It's a faster, cheaper way to coax out a signal. It also usually pairs with a smaller flip angle (tipping the protons only partway over instead of a full 90°), which means you don't have to wait as long for the tissue to recover before the next pulse. Faster pulses, no slow refocusing step — the result is speed. This is what lets you hold your breath once and image the whole abdomen, or watch the heart beat in near-real time.
But there's no free lunch. The gradients can reverse the dephasing they caused, but they cannot undo the dephasing from the magnet's own imperfections — there's no runner U-turn to swap the head starts. So all that static field unevenness stays baked into the signal.
Spin echo signal decays by true T2. Gradient echo signal decays by T2* ("T2-star") — T2 plus all the extra dephasing from field imperfections that the 180° pulse would have erased. T2* is always faster (shorter) than T2.
The same weakness is also the superpower
Now, the fun part. That sensitivity to field distortion isn't only a bug — it's exactly what you want when the thing you're hunting creates a field distortion.
Blood breakdown products, calcium, air, and metal all warp the local magnetic field around themselves (we call this susceptibility). Gradient echo, having no 180° pulse to clean things up, lights these up as blooming black spots. That's why gradient echo — and its souped-up cousins — are the workhorses for finding tiny hemorrhage and susceptibility effects, like microbleeds in the brain.
"Blooming" cuts both ways. Gradient echo exaggerates the size of metal, calcium, and old blood, so a speck of hemosiderin can look like a crater. Useful for detecting it — misleading if you try to measure it. And near big metal hardware or air-tissue interfaces, gradient echo can drop signal so badly the anatomy vanishes. When metal is in the picture, spin echo is your friend.
Putting them side by side
| Spin echo | Gradient echo | |
|---|---|---|
| Refocusing pulse | Yes — 180° | No |
| Signal decays by | True T2 | T2* (faster) |
| Speed | Slower | Faster |
| Field imperfections | Corrected, ignored | Preserved, emphasized |
| Susceptibility / metal | Less artifact (good for hardware) | Blooms (good for blood, bad near metal) |
| Typical role | Routine anatomy workhorse | Fast imaging, breath-holds, hemorrhage hunting |
So the one-sentence summary I'd tattoo on the inside of my eyelids: spin echo spends time to ignore the magnet's flaws; gradient echo saves time by embracing them — and that embrace is exactly why it's so good at spotting blood and so bad near metal. Everything else is detail. If you want to see how those gradients actually build the image line by line, that's the story over in gradients and k-space.