How to Reduce Motion Artifacts in MRI: A Technologist's Checklist

You set up a perfect knee, the localizer looks great, and three minutes into a long sequence the patient shifts their leg. The images come back with ripples marching across the joint, and now you are deciding whether to repeat, move on, or change the plan. Every MRI technologist knows that moment.

Motion cannot be eliminated completely, but most of it can be prevented, and much of the rest can be managed with tools already on your scanner. This guide explains why motion shows up the way it does, then gives you a scanner-side checklist to use before the exam and at the first sign of trouble.

Why motion looks the way it does

MRI builds an image from many lines of k-space collected over time. If the anatomy moves between those lines, the data no longer agree with each other. With conventional line-by-line (Cartesian) sampling, that inconsistency appears as blurring and as ghosts, and the ghosts run along the phase-encoding direction no matter which way the patient actually moved.

Two practical points follow from that:

• Periodic motion, such as breathing and arterial pulsation, is predictable. It responds well to gating, triggering, fast acquisitions and radial sampling.
• Random motion, such as bowel peristalsis or a patient shifting position, is not predictable. The best fix is usually to remove the cause instead of trying to out-engineer it.

Knowing which kind you are looking at tells you which tool to reach for.

Step 1: Prevent motion before the first sequence

Prevention is the least expensive fix you have, and almost all of it happens before you leave the scan room.

Coach the patient

A 2024 RadioGraphics review of motion in abdominal MRI, and RSNA's coverage of it, put coaching the patient before and during the exam at the top of the list. In practice:

• Explain what is about to happen in plain words: how long the exam is, that the scanner is loud, and that the noise is the signal to hold still.
• Tell patients when they can move. People hold still better when they know a break is coming.
• Practice breath-holds before scanning. A brief breath-hold check on an alert patient outside the scan room shows you early whether a breath-hold protocol is realistic, so you can choose the right approach from the start.
• Give the same breathing instruction, in the same words, every time, and ask for regular, steady breathing between holds.
• Keep talking between sequences. Tell the patient how long the next one is and how they are doing.

Make holding still easy

A comfortable patient moves less. Take the extra minute to:

• Support the part being scanned with foam pads and positioning aids, so the patient is resting instead of holding a position with muscle effort.
• Use straps, tape or a snug sheet wrap where appropriate. Infants are commonly swaddled with the head supported.
• Deal with the obvious distractions first: a bathroom stop, a blanket, a bolster under the knees, properly fitted hearing protection.
• Ask about pain. A patient who cannot lie flat for long needs a shorter plan, not more encouragement.

Use only padding and immobilization devices approved for your MR environment, and follow your facility's positioning policies.

Know when coaching is not enough

Some patients cannot hold still however well you prepare them, including young children, patients in significant pain, and patients who are anxious or confused. Sedation or anesthesia may be required, and medication that slows peristalsis is sometimes used for bowel motion. Those are decisions for the radiologist under your facility's protocols, not something to solve at the console. Your part is to flag the problem early, before table time is lost.

Step 2: Remove the signal from tissue that moves

If moving tissue produces no signal, it cannot ghost across the anatomy you care about.

• Saturation bands. Place a spatial saturation pulse over the moving structure, such as the anterior abdominal wall on a lumbar spine exam.
• Fat suppression. Bright fat near moving structures is a frequent source of motion artifact. Suppressing it removes much of the artifact along with it.
• Flow saturation and flow compensation. Saturation pulses can null blood entering the slice, and flow compensation (gradient moment nulling) reduces ghosting from pulsatile blood and CSF.
• Coil choice. A surface coil limited to the area of interest picks up less signal from distant moving tissue.

Step 3: Change how the data are collected

When motion is still getting through, adjust the acquisition.

• Swap phase and frequency. This does not reduce the artifact. It moves it. That is still worth doing when it shifts ghosts off the anatomy of interest. Check for wrap after you swap.
• Go faster. Single-shot and other ultrafast sequences can freeze bulk motion, at some cost in resolution or contrast.
• Add averages. More signal averages reduce motion artifact and improve SNR, but they lengthen the scan, which can backfire with a patient who is already struggling.
• Use radial or blade-style sampling. Radial trajectories spread motion out as mild blurring or streaks instead of discrete ghosts, which is generally less objectionable. Blade-style sequences oversample the center of k-space and can correct for in-plane motion. Names differ by manufacturer, so check what your system calls them.
• Gate or trigger. Respiratory bellows or belts, ECG or pulse sensors, and navigator echoes let the scanner collect data at a consistent point in the cycle. Expect longer scan times when breathing is irregular.

Step 4: Have a rescue plan

The RadioGraphics review describes a practical way to handle breath-hold failure in abdominal imaging:

• If breath-holding fails late in the exam, add a free-breathing radial sequence to recover the key images.
• If it deteriorates early, switch to free-breathing imaging for the rest of the exam.
• If you expect a poor breath-holder from the start, begin with a free-breathing protocol.

The authors note that a free-breathing radial acquisition is often faster than several failed breath-hold attempts. The habit carries over to other exams: repeating the same sequence the same way on the same patient rarely gives a different result, so change something. Any protocol change should stay within what your radiologists have approved.

Quick checklist for the scan room

1. Explain the exam and the "noise means hold still" rule.
2. Practice the breath-hold and decide whether it is realistic.
3. Pad, support and secure. Check comfort and pain.
4. Run the sequences that matter most to the clinical question early, as your protocol allows.
5. Check the first images for motion before moving on.
6. If motion appears, identify it: periodic or random?
7. Pick the matching tool: saturation, fat suppression, flow compensation, phase swap, a faster or radial sequence, or gating.
8. Do not repeat unchanged. Re-coach, re-pad or change the technique.
9. Document what you tried so the radiologist knows.

Key takeaways

• Motion ghosts appear in the phase-encoding direction regardless of the direction of movement.
• Coaching and comfort come first. They cost nothing in scan time and prevent motion at the source.
• Periodic motion responds to gating, fast imaging and radial sampling. Random motion is best handled by removing the cause.
• Swapping phase and frequency relocates an artifact. It does not remove it.
• Radial and blade-style sequences trade ghosts for mild blurring or streaks.
• Have a rescue plan and switch early instead of repeating the same failed sequence.

This article is educational and is not patient-specific medical advice. Always follow your facility's policies and protocols, the manufacturer's instructions for your system and accessories, and the direction of your radiologist, MR Safety Officer and medical physicist.

Sources

• T1-weighted Motion Mitigation in Abdominal MRI: Technical Principles, Clinical Applications, Current Limitations, and Future Prospects (RadioGraphics, 2024)
• Radiologists Can Deftly Curb Motion Artifacts in Abdominal MRI (RSNA News, January 2025)
• Basic Theory of Motion Artifacts in MRI (ISMRM Proceedings, 2021)
• Reducing Motion Artifacts (MRIquestions.com, Allen D. Elster, MD)