This step-by-step guide is for MRI students, radiographers, and technologists who wish to improve their planning skills and master the MRCP protocol.
What you will learn:
Key factors in MRCP, including trade-offs.
Patient and scanner setup tips.
Best pulse sequences and planning techniques.
Ways to avoid common artifacts.
What great MRCP images should look like.
Key Takeaways
Prioritize resolution, then SNR, then scan time, but nothing matters until you control motion.
The bile and pancreatic ducts are fluid-filled tubes only a few millimeters wide, so small stones and strictures disappear into partial volume blurring without high resolution. Strong SNR keeps bright fluid clearly separated from dark stones. But the upper abdomen moves constantly with breathing, so respiratory motion is the biggest practical threat to image quality.
The protocol splits into two groups: one built for detail, one built for speed.
Respiratory-triggered sequences like the 3D MRCP spend time to keep resolution and SNR high. Breath-hold sequences like the single-shot and radial thick-slab acquisitions put speed first and give up some detail to freeze motion. Together they keep the study diagnostic no matter how well the patient copes.
Avoid these 5 common MRCP artifacts.
Artifact
Solution – How to Avoid It
Respiratory motion
Coach the patient on calm, regular breathing before the scan. Use respiratory or navigator-triggered 3D MRCP when available. If the 3D sequence is degraded, fall back on fast 2D thick-slab MRCP as a motion-resistant backup.
Susceptibility (gas, clips, pneumobilia)
Have the patient fast for 4–6 hours when protocol allows. Cross-check any dark focus against SS-TSE/HASTE images and the 3D MRCP source images. Review for pneumobilia, clips, stents, or prior sphincterotomy before calling something a stone.
Review the finding across multiple planes and on the source images. A true stone stays within the duct on every slice. A pulsation artifact usually lines up with an adjacent vessel, most often the hepatic artery near the common hepatic duct or mid-CBD.
Partial volume / MIP averaging
Never rely on the thick MIP image alone. Always review the thin 3D source images, since MIP reconstruction can hide small stones or exaggerate strictures and blur.
GI fluid overlap
Fast beforehand. Use a site-approved negative oral contrast agent when local protocol allows it, such as pineapple or blueberry juice, to darken bright stomach and duodenal fluid that can overlap the bile duct.
Intro to the MRCP Protocol
MRCP, or magnetic resonance cholangiopancreatography, evaluates the biliary and pancreatic duct systems using heavily T2-weighted imaging. This makes static fluids like bile and pancreatic juice appear bright, without invasive procedures or ionizing radiation. Stones contain no free fluid, so they appear as dark filling defects against that bright background.
The protocol combines routine upper-abdominal sequences with dedicated MRCP acquisitions, so we assess both the abdominal organs and the fine duct anatomy in one study.
MRCP is now the standard non-invasive test for suspected common bile duct stones and biliary obstruction, with reported sensitivity generally in the 85–100% range and specificity around 90–100%. ERCP is increasingly reserved for cases that need therapy rather than diagnosis.
In MRI, we always face a trade-off between 3 key metrics:
Scan Time: How fast a pulse sequence can be completed.
Resolution: How much detail the image can display.
SNR: How clear the image is, how much signal relative to noise.
Improving one of these metrics reduces the performance of the others. To decide what trade-offs to make, we must consider the needs of each clinical situation.
For MRCP, we face three specific challenges:
Tiny structures. The main pancreatic duct is usually only a few millimeters wide, up to about 3 mm on MRCP depending on location and age, and stones or strictures can be smaller still. Without high spatial resolution, partial volume averaging blurs them together and small findings vanish.
Fluid-based contrast. The duct signal comes mainly from fluid. Heavily T2-weighted imaging keeps static bile and pancreatic juice bright against a darker background, so we need enough SNR to hold the bright-fluid to dark-stone distinction clear. That gets SNR-sensitive when voxels are small and echo times are long.
Constant motion. The upper abdomen moves with breathing and bowel motion, and the high-detail scan takes minutes.
Therefore, we typically:
Prioritize resolution because the fine duct detail is the diagnosis, and
Keep SNR high enough for clarity.
But neither matters until motion is controlled, and one scan usually cannot maximize resolution, SNR, and speed at once.
So the protocol takes a two-pronged approach, splitting its sequences into two groups: one optimized for detail, one for speed.
1. The respiratory-triggered sequence group: 1) resolution, 2) SNR, 3) scan time
These sequences acquire data during a consistent part of the breathing cycle, usually near end-expiration, which reduces motion rather than freezing it, at the cost of minutes. Having already spent the time, they keep resolution and SNR high, so scan time comes last.
The 3D MRCP is the most important sequence. Its thin near-isotropic voxels, roughly 1.4 mm in-plane by 1.6 mm slice, resolve small stones and fine ducts and reconstruct in any plane, and its volumetric readout supplies the SNR those thin slices spend. The coronal and fat-suppressed T2 images ride along in the same group, trading time for clean, detailed anatomy.
2. The breath-hold sequence group: 1) scan time, 2) SNR, 3) resolution
These sequences finish inside a single breath-hold or a single shot, so respiratory motion has little chance to blur them, though bowel motion or a failed breath-hold still can.
Speed comes first, and detail is what gets sacrificed. The thick slab collapses through-plane detail into one projection, and single-shot readouts come out softer than the triggered images. This group keeps the study diagnostic when the patient cannot hold still or breathe regularly.
Strategy and Priority for Each MRCP Sequence
This table shows a breakdown of every sequence in standard MRCP protocol, the motion-control strategy we use, and what to prioritize in acquisition.
Note! Prioritizing resolution in MRCP is only a general guideline, NOT a strict rule. If your patient cannot breathe regularly or hold still, the fastest diagnostic image beats the sharpest non-diagnostic one. The right balance always depends on the needs of your patient and clinic.
Note on Other MRCP Variants
Grouping and triggering differ by vendor and site, so a sequence that is respiratory-triggered here may be run breath-hold elsewhere.
Some centers use breath-hold 3D MRCP with compressed sensing to bring near-3D duct detail into one breath-hold, though heavy acceleration can reduce small-duct visibility.
Secretin MRCP changes the trade-off further, adding repeated timed thick-slab images after IV secretin to assess pancreatic duct filling and function, where temporal timing outranks spatial resolution. These are not the sequences demonstrated in this video.
MRCP Health Conditions and the MRI Sequences That Reveal Them
The MRCP study can help us diagnose a wide range of health conditions. The table below lists some of the most common conditions and what pulse sequences reveal them:
Common conditions
Clearly seen on sequence
Why this sequence?
• Rough level of biliary obstruction
• Large extra-biliary masses
Cor T2
A multi-shot T2 TSE covering the whole upper abdomen along the axis of the biliary tree. Good for orientation and catching large findings, though the exact obstruction level and fine duct detail are confirmed on 3D MRCP.
Fat suppression removes bright background fat and the multi-shot readout adds resolution, so inflammatory fluid, wall thickening, and edema show up where a faster sequence would blur them.
• Possible CBD stones as dark filling defects
• Biliary dilatation
Ax T2 SS-TSE
Freezes a slice in about one second, resisting motion, which is its main strength. This makes it the go-to quick duct screen when breath-holding is poor, though flow artifact can mimic a filling defect, so suspected stones are confirmed on 3D MRCP source images.
Separates fat and water signal into four image sets, telling fat or blood apart from a true lesion. Useful for an iron or susceptibility clue too, though formal iron quantification needs a dedicated multi-echo method. Not the main sequence for finding stones.
• CBD stones and biliary obstruction
• Strictures: PSC or cholangiocarcinoma pattern
• Duct variants and IPMN communication
3D MRCP
The main duct-map sequence. Ultra-long echo times leave mostly static fluid visible, and thin isotropic voxels reconstruct the duct system in any plane. A stricture here shows the narrowing pattern, not full tumor staging, which needs contrast T1, DWI, or CT/EUS. Very small stones can still be missed, so source images are always checked alongside the MIP.
• CBD stones and biliary obstruction
• Ductal dilatation
Radial SS-TSE FS
Each thick slab gives a motion-resistant projection image in about one to two seconds, so it's the backup when 3D MRCP is degraded by motion. It shows the same duct pathology as the 3D, just faster and coarser.
How to Perform an MRCP
The step-by-step guide below will show you how to set up and perform an MRCP protocol in practice.
We will perform the protocol in 3 parts:
Set up the Patient and MRI Scanner
Plan and Acquire the Protocol Sequences
Review the Images
Part 1: Set up the Patient and MRI Scanner
1. Prepare the Patient
MRCP is one of the few protocols where preparation before the patient reaches the table directly affects image quality.
Fasting: Ask the patient to fast for 4–6 hours when the clinical situation allows. This reduces bright fluid in the stomach and duodenum, and lets the gallbladder distend.
Negative oral contrast (optional, site-dependent): Bright gastric and duodenal fluid can sit on top of the common bile duct and hide it. Commercial ferumoxsil agents are purpose-built for this. Many departments use pineapple juice instead, which works because its naturally high manganese content shortens the T2 of the fluid it mixes with, so at MRCP’s very long echo times that fluid loses nearly all signal. Published protocols typically give 200–400 mL, 10–30 minutes before scanning. Always follow your own institution’s policy.
Breathing coaching: Practice breath-hold instructions before the patient enters the bore, and test both exhale and inhale. Many patients hold their breath far more reliably one way, and picking the better one keeps slice positions consistent between sequences.
2. Position the Patient in the Scanner
Lay the patient supine with the upper abdomen centered at the scanner’s isocenter.
Use an anterior body array coil on top of the upper abdomen, combined with the posterior spine coil integrated into the scanner table. This gives full signal coverage of the liver, gallbladder, biliary tree, and pancreas.
Arms can be positioned above the head or alongside the body. If they rest alongside the body they sit near the field of view in the right-to-left direction, which is why fold-over suppression matters on the coronal sequences.
Once the patient is in place, review your scanner’s hardware settings.
In this guide, we will use the following settings:
Scanner Setting
Value
Why This Value
Magnetic field strength
1.5 T
Provides good field homogeneity and image quality for abdominal imaging while staying less prone to susceptibility and dielectric artifacts in the abdomen than 3 T.
Maximum gradient strength
45 mT/m
Enables faster acquisitions while preserving high image quality.
Receive coil
Anterior body array + posterior spine coil
Full anterior-to-posterior coverage of the upper abdomen with strong signal reception.
This hardware setup is widely used in clinical practice. It balances acquisition time, image quality, and patient comfort.
4. Capture the Initial Localizer Images
Before we can perform any MRI protocol, we must always capture initial localizer images of the patient. These images act as a guide for planning the detailed scans we will perform next.
Select the abdominal region, save it, and run the localizers.
We should always capture localizers in three planes:
Axial
Sagittal
Coronal
Once acquired, drag and drop the localizer images into the three viewports. Then scroll through each stack to locate a central slice that clearly shows the anatomy of the upper abdomen.
✅ Correct Setup of Localizer Images for MRCP:
Part 2: Plan and Acquire the Protocol Sequences
When all preparations are ready, we can start planning and acquiring the protocol sequences.
Let’s go through the pulse sequences a standard MRCP protocol includes, why we perform them, and how to set them up.
The 6 Sequences of a Standard MRCP Protocol
Coronal T2 TSE
Axial T2 FS TSE
Axial T2 SS-TSE
Axial GRE Dixon
3D MRCP
Radial SS-TSE FS
Sequences 1, 2, and 5 belong to the respiratory-triggered group. Sequences 3, 4, and 6 belong to the breath-hold group.
Note! Your facility may have its own local policies, so the protocol might differ in some of these sequences. If they differ, then always follow your own institution's guidelines.
In the sections below, we go through how to plan and set up each sequence.
1. Planning Coronal T2 TSE
✅ Correct Planning:
Planning Instructions:
Plan in the true coronal plane, covering the entire hepatobiliary system.
Use appropriate geometry parameters:
Coverage: From the anterior abdominal wall back to the kidneys, including liver, gallbladder, pancreas, spleen, and upper kidneys.
Slice number: 32–38, enough for that anterior-to-posterior block.
Slice thickness: 6 mm, standard overview thickness.
Slice gap: 1 mm, prevents crosstalk while maintaining continuity.
Set the fold-over direction (phase encoding) to right-to-left (RL) and activate fold-over suppression, so the arms and body wall do not alias into the image.
Acquire with respiratory triggering. This is a multi-shot TSE that takes a few minutes, so it follows the breathing cycle rather than a breath-hold.
Tip: You can display your slices in either box view or slice view. Pick whichever helps you check coverage, angle, and position most clearly.
Parameters for Coronal T2 TSE:
Parameter
Recommended Values
Why These Values
Effective Echo Time (TE)
80–100 ms
Long TE is required for T2 contrast.
Repetition Time (TR)
2,000–2,500 ms
Long TR is required for T2 contrast. With respiratory triggering, the effective TR follows the breathing cycle.
Field-of-View (FOV)
380 × 380 mm
Covers the full width of the upper abdomen at this patient size.
Matrix
320 × 224
High read matrix for in-plane detail, reduced phase matrix to hold scan time down on a multi-shot readout.
Foldover Direction (Phase)
Right-to-Left (RL)
Matches the shape of the coronal upper abdomen.
Number of Slices
32–38
Enough to cover from the anterior abdominal wall back to the kidneys.
Slice Thickness
6 mm
Standard overview thickness, keeps SNR high without excessive scan time.
Slice Gap
1 mm
Prevents crosstalk between slices while maintaining continuity.
NEX / Averages
1
Keep as low as possible to hold scan time reasonable on a triggered acquisition.
Bandwidth
220–250 Hz/px
Just above the point where fat-water shift equals one pixel at 1.5 T. Going lower would gain SNR but widen chemical shift at the organ borders.
Turbo Factor / ETL
15–25
A high turbo factor collects many echoes per shot, which is what keeps a multi-shot readout of 35 slices down to a few minutes.
Fold-over Suppression
Yes
Prevents aliasing from the arms and body wall in the RL direction.
Breath-hold Mode
Free breathing
Multi-shot TSE takes minutes, which is far too long for a breath-hold, so it follows the respiratory cycle instead.
2. Planning Axial T2 FS TSE
✅ Correct Planning:
Planning Instructions:
Plan in the axial plane. Right-click on the axial localizer and set the in-plane position, which also sets the correct orientation under the Geometry tab.
Use appropriate geometry parameters:
Coverage: From the hepatic dome down to the kidneys.
Slice number: 32–38.
Slice thickness: 6 mm, with a 1 mm gap.
Review your slice package on top of a high-resolution image, such as the coronal T2, and scroll through to confirm the abdominal dome and kidneys are both covered.
Set the fold-over direction (phase encoding) to anterior-to-posterior (AP) to prevent aliasing.
Fold-over suppression can stay off here, since no anatomy extends beyond the field of view in the AP direction. Leaving it off also reduces scan time.
Acquire with respiratory triggering, the same as the coronal T2.
Parameters for Axial T2 FS TSE:
Parameter
Recommended Values
Why These Values
Effective Echo Time (TE)
80–95 ms
Long TE is required for T2 contrast.
Repetition Time (TR)
2,000–2,500 ms
Long TR is required for T2 contrast and allows enough slices per acquisition.
Field-of-View (FOV)
380 × 380 mm
Sized to the patient's abdominal width, matching the coronal T2.
Matrix
320 × 256
Higher phase matrix than the coronal T2, resolving wall thickening and small fluid collections.
Foldover Direction (Phase)
Anterior-to-Posterior (AP)
Matches the shape of the axial abdomen and prevents wraparound.
Number of Slices
32–38
Enough to cover from the hepatic dome to the kidneys.
Slice Thickness
6 mm
Standard thickness for an axial abdominal survey.
Slice Gap
1 mm
Prevents crosstalk while maintaining continuity.
NEX / Averages
2
Higher than the coronal T2, since the fat suppression pulse removes signal that has to be recovered.
Bandwidth
220–250 Hz/px
Kept low to protect the SNR that the fat suppression pulse removes, while staying high enough to hold chemical shift to about one pixel at 1.5 T.
Turbo Factor / ETL
18
Matched to the 80 ms effective echo time to keep acquisition time suitable.
Fold-over Suppression
No
No anatomy extends beyond the FOV in the AP direction, and leaving it off shortens scan time.
Fat Suppression
Spectral
Removes fat signal so inflammation and edema stand out against a darker background.
Breath-hold Mode
Free breathing
Multi-shot TSE needs minutes to acquire, so it follows the respiratory cycle instead of a breath-hold.
3. Planning Axial T2 SS-TSE
The SS tag stands for single-shot. This sequence captures the complete k-space for each slice in one shot, which makes it far faster than the multi-shot sequences before it.
✅ Correct Planning:
Planning Instructions:
Copy the slice geometry and planning from the previous axial T2 FS TSE sequence.
Keep the same slice angulation, coverage, and positioning so images of different contrasts can be clearly compared.
Acquire as a breath-hold, at exhale or inhale depending on which your patient performs more reliably.
Parameters for Axial T2 SS-TSE:
Parameter
Recommended Values
Why These Values
Effective Echo Time (TE)
120–140 ms
Higher TE than the axial T2 FS gives better fluid-to-soft-tissue contrast, leaning toward highlighting fluid-containing structures.
Repetition Time (TR)
1,000–1,500 ms
Each slice is a single shot, so TR sets the interval between slices rather than the contrast.
Field-of-View (FOV)
380 × 380 mm
Copied from the axial T2 FS for direct comparison.
Matrix
320 × 256
Copied from the axial T2 FS. The softer look comes from the single-shot readout and half-Fourier filling, not from a smaller matrix.
Foldover Direction (Phase)
Anterior-to-Posterior (AP)
Matches the axial anatomy shape.
Number of Slices
32–38
Copied from the axial T2 FS.
Slice Thickness
6 mm
Copied from the axial T2 FS.
Slice Gap
1 mm
Copied from the axial T2 FS.
NEX / Averages
1
Single average, since the point of this sequence is speed.
Bandwidth
600–700 Hz/px
Roughly three times the 1.5 T minimum. Wide bandwidth shortens the readout and echo spacing, which is what lets a whole slice be read in one shot.
Partial Fourier
Phase-conjugate (half-Fourier)
Fills only part of k-space and reconstructs the rest, which is what enables the single-shot readout.
Fold-over Suppression
No
Copied from the axial T2 FS, keeps acquisition fast.
Breath-hold Mode
Exhale or inhale
Each slice freezes in roughly one second, so the stack fits inside one or two breath-holds. Pick whichever your patient holds more reliably.
4. Planning Axial GRE Dixon
✅ Correct Planning:
Planning Instructions:
Copy the slice geometry from one of your previous axial sequences using the Custom copy option.
This time, remove the matrix sizes from the copy while keeping everything else. The Dixon runs at slightly lower in-plane resolution because it is acquired in a breath-hold.
Through-plane resolution and central position stay the same, so the radiologist can compare all sequences at matching levels.
Acquire as a breath-hold.
Parameters for Axial GRE Dixon:
Parameter
Recommended Values
Why These Values
Echo Time (TE)
4.5 ms
At 1.5 T, fat and water return to the in-phase condition every 2.24 ms, so 4.5 ms lands on an in-phase echo.
Second-echo Time Difference
2.24 ms
Half a fat-water phase cycle at 1.5 T, placing the second echo at 6.74 ms, which is opposed-phase. The two echoes are what make the Dixon separation possible.
Repetition Time (TR)
150–200 ms
Short TR for a fast T1-weighted gradient echo readout.
Flip Angle
15°
Low flip angle suited to the short TR.
Field-of-View (FOV)
380 × 380 mm
Copied from the axial T2 sequences.
Matrix
320 × 224
Reduced phase matrix compared with the axial T2 sequences, which is the resolution cost of fitting the acquisition into one breath-hold.
Foldover Direction (Phase)
Anterior-to-Posterior (AP)
Matches the axial anatomy shape.
Number of Slices
32–38
Copied from the axial T2 sequences.
Slice Thickness
6 mm
Keeps through-plane resolution equivalent to the other axials.
Slice Gap
1 mm
Copied from the axial T2 sequences.
NEX / Averages
1
Single average keeps scan time short enough for one breath-hold.
Bandwidth
600–700 Hz/px
Wide bandwidth is what allows both echoes to be read at the exact timings the Dixon method needs, and it keeps chemical shift well under a pixel.
Fold-over Suppression
No
No anatomy extends beyond the FOV in the AP direction.
Fat Suppression
None (reconstructed)
Fat and water are separated mathematically from the two echoes, not by a suppression pulse.
Breath-hold Mode
Exhale or inhale
Fast gradient echo readout completes inside a single breath-hold. Match the direction used on the single-shot to keep slice positions consistent.
How Dixon works: The sequence acquires two echoes, one where fat and water signals are in phase and one where they are opposed. From those, the scanner reconstructs four image sets: in-phase, out-of-phase, water-only, and fat-only.
5. Planning 3D MRCP
This is the cornerstone of the study. Everything before it establishes anatomy and context. This sequence maps the duct tree itself.
✅ Correct Planning:
Planning Instructions:
Bring the sagittal localizer back into a viewport and scroll to a slice showing the upper-abdomen organs clearly. Keep your high-resolution coronal and axial images ready for planning.
Plan the slab as a coronal. Position on the coronal T2 image, and set the in-plane position. The slab should then sit perpendicular on the axial view.
Use the axial T2 FS to check the extent of the gallbladder, then scroll down to find the choledochus and pancreatic duct. These are the structures you must always cover.
Use appropriate geometry parameters:
Slice number: 75–85, enough to cover the hepatobiliary system and no more. Every extra slice costs time on an already long sequence.
Slice thickness: 1.6 mm, thin near-isotropic voxels.
Slice gap: 0 mm, contiguous slices are required for a true 3D volume.
Angle the slab strategically so it captures those structures without including surrounding anatomy that adds nothing here.
Acquire with respiratory triggering. Verify the patient’s breathing is consistent, otherwise the listed scan time will stretch out significantly.
Parameters for 3D MRCP:
Parameter
Recommended Values
Why These Values
Effective Echo Time (TE)
500–800 ms
Ultra-long TE for heavy T2 weighting, so only static fluid retains signal. Higher TE means lower signal, so balance against SNR.
Repetition Time (TR)
2,500–3,000 ms
Long TR gives enough relaxation for the heavily T2-weighted readout. With respiratory triggering, the effective TR follows the breathing cycle.
Field-of-View (FOV)
340 × 340 mm
Focused on the hepatobiliary region rather than the whole abdomen.
Matrix
244 × 244
Combined with the FOV, gives roughly 1.4 mm isotropic in-plane voxels.
Foldover Direction (Phase)
Foot-to-Head (FH)
Runs along the long axis of the coronal slab, which keeps any residual aliasing outside the duct region.
Number of Slices
75–85
Enough to cover the biliary tree and pancreatic duct, and no more.
Slice Thickness
1.6 mm
Thin near-isotropic voxels resolve small stones and fine ducts, and allow reconstruction in any plane.
Slice Gap
0 mm
Contiguous slices are required for a true 3D volume and clean reconstructions.
NEX / Averages
1
Single average, since the sequence is already long.
Bandwidth
400–450 Hz/px
Short echo spacing matters across a 150-echo train, so bandwidth sits well above the 1.5 T minimum even though that costs SNR.
Turbo Factor / ETL
150
Very long echo train collects the whole slice efficiently within each respiratory window.
Parallel Imaging
No
Not used here, since parallel imaging costs SNR the thin slices cannot spare.
Partial Fourier
Phase-conjugate (half-Fourier) Factor 0.6
Fills 60% of k-space in the phase direction and reconstructs the rest, which is the acceleration used instead of parallel imaging.
Fold-over Suppression
Yes
Prevents wraparound in the FH direction.
Fat Suppression
Spectral
Suppresses background signal so only the fluid-filled ducts appear bright.
Breath-hold Mode
Free breathing
Data is collected at a consistent point in each breath, usually near end-exhalation, rather than frozen in a breath-hold.
6. Planning Radial SS-TSE FS
This is a thick-slab, projection-style acquisition. Instead of a stack of parallel slabs, it rotates a small number of thick slices around a single central axis.
✅ Correct Planning:
Planning Instructions:
Bring back a high-resolution coronal T2 image for planning.
Set the first slab perpendicular on the axial image and center it. That slab sits in a true coronal plane through the biliary tree.
Enable radial slices, and set the number of slices.
The slabs rotate at a fixed angle around that initial position. Completing 360° gives multiple projection views of the duct tree from different angles.
Typically 9–12 slabs are enough, since each views the ducts from a different angle.
Acquire as a breath-hold at exhale, matching the single-shot sequence so slice positions stay consistent.
Parameters for Radial SS-TSE FS:
Parameter
Recommended Values
Why These Values
Effective Echo Time (TE)
800–1,000 ms
Very high TE for a heavily T2-weighted projection image, similar contrast to the 3D MRCP.
Repetition Time (TR)
5,000–6,000 ms
Long TR gives full relaxation between slabs for the heavily T2-weighted single-shot readout.
Field-of-View (FOV)
380 × 380 mm
Wide enough to keep the whole duct system inside every rotated slab.
Matrix
288 × 288
Square matrix gives roughly 1.3 mm isotropic in-plane resolution, which is needed because the slab rotates.
Foldover Direction (Phase)
Right-to-Left (RL)
Matches the coronal slab orientation.
Slice Thickness
40–45 mm
Thick slab produces a projection-style image of the whole duct system, similar to conventional cholangiography.
Number of Radial Slabs
9–12
Each slab views the duct tree from a different angle through a full 360° rotation.
NEX / Averages
1
Single average, since each slab completes in roughly one to two seconds.
Bandwidth
400–450 Hz/px
Short echo spacing is what allows a single-shot readout at this echo time, so bandwidth sits well above the 1.5 T minimum.
Use Minimum Echo Spacing
Yes
Packs the echo train as tightly as possible so each slab finishes in seconds.
Parallel Imaging
No
Not used here. The speed comes from the single-shot readout and half-Fourier instead.
Partial Fourier
Phase-conjugate (half-Fourier) Factor 0.6
Fills 60% of k-space in the phase direction and reconstructs the rest, which is what makes a single shot possible at this echo time. Same approach as the 3D MRCP.
Fold-over Suppression
Yes
Prevents wraparound across the wide rotating field of view.
Fat Suppression
Spectral
Suppresses background fat so the bright ducts dominate the projection.
Breath-hold Mode
Exhale
The whole radial set finishes in a few seconds, and matching the single-shot breath-hold keeps slice positions consistent.
How to Avoid Artifacts When Planning the Sequences
The table below lists the 5 common MRCP artifacts, and what techniques you can use to avoid them:
Artifact
Solution – How to Avoid It
Respiratory motion
Coach the patient on calm, regular breathing before the scan. Use respiratory or navigator-triggered 3D MRCP when available. If the 3D sequence is degraded, fall back on fast 2D thick-slab MRCP as a motion-resistant backup.
Susceptibility (gas, clips, pneumobilia)
Have the patient fast for 4–6 hours when protocol allows. Cross-check any dark focus against SS-TSE/HASTE images and the 3D MRCP source images. Review for pneumobilia, clips, stents, or prior sphincterotomy before calling something a stone.
Review the finding across multiple planes and on the source images. A true stone stays within the duct on every slice. A pulsation artifact usually lines up with an adjacent vessel, most often the hepatic artery near the common hepatic duct or mid-CBD.
Partial volume / MIP averaging
Never rely on the thick MIP image alone. Always review the thin 3D source images, since MIP reconstruction can hide small stones or exaggerate strictures and blur.
GI fluid overlap
Fast beforehand. Use a site-approved negative oral contrast agent when local protocol allows it, such as pineapple or blueberry juice, to darken bright stomach and duodenal fluid that can overlap the bile duct.
Part 3: Review the Images
Finally, we will review the images to ensure all the anatomical information we need is clear.
These key structures must be clearly visible in an MRCP:
Intrahepatic bile ducts, right and left hepatic ducts
Common hepatic duct and cystic duct
Gallbladder, including wall and contents
Common bile duct along its full length to the ampulla
Main pancreatic duct
Pancreatic parenchyma
Liver, spleen, and upper kidneys
Below, we will go through all the different image contrasts and explain their specific role in an MRCP study.
Coronal T2 TSE – Anatomical Overview of the Upper Abdomen
The coronal T2 gives a high-resolution anatomical overview of the upper abdomen, showing the liver, pancreas, gallbladder, biliary tree, and surrounding structures in a plane that runs along the axis of the biliary system.
It is valuable for detecting edema, inflammation, masses, and fluid collections, and for establishing anatomical orientation before the dedicated MRCP acquisitions.
✅ Coronal T2 TSE – Correct Image Example:
Things to Look for in Coronal T2:
Coverage runs correctly from anterior to posterior, capturing liver, gallbladder, pancreas, spleen, and upper kidneys.
The biliary tree can be traced across its general course.
No aliasing from the arms or body wall in the right-to-left direction.
Large extra-biliary masses and fluid collections are visible.
Axial T2 FS TSE – Inflammation and Edema
Suppressing the fat increases the visibility of fluid, edema, and inflammatory changes, while improving lesion detection within the liver. This sequence is particularly useful for evaluating pancreatitis, cholangitis, fluid collections, and cystic lesions, as well as gallbladder wall thickening in cholecystitis.
✅ Axial T2 FS TSE – Correct Image Example:
Things to Look for in Axial T2 FS:
Fat is uniformly suppressed across the whole field of view.
Coverage runs from the hepatic dome down to the kidneys.
Gallbladder wall thickness and any pericholecystic fluid are clearly assessable.
Peripancreatic fluid and edema stand out against the darker background.
Axial T2 SS-TSE – Fast Motion-Resistant Duct Screen
The axial T2 single-shot is a reliable overview sequence, helping identify free fluid, biliary dilatation, pancreatic abnormalities, and other upper-abdominal pathology. Because it is planned identically to the axial T2 FS, the two can be compared slice for slice.
✅ Axial T2 SS-TSE – Correct Image Example:
Things to Look for in Axial T2 SS-TSE:
Slice positions match the axial T2 FS exactly.
Images look slightly softer even though the matrix matches the axial T2 FS. That blurring comes from the long single-shot echo train and half-Fourier filling, and is the expected cost of speed.
Contrast on the vital structures remains well-defined.
Dark filling defects in the CBD are noted but confirmed on 3D MRCP source images, since flow artifact can mimic a stone.
Axial GRE Dixon – Tissue Characterization
The Dixon sequence gives tissue characterization by separating fat and water signal. It is useful for detecting hepatic steatosis, microscopic fat, adrenal adenomas, pancreatic fat infiltration, and hemorrhage, while producing high-quality fat-suppressed images. The top-left corner of the viewport shows which of the four reconstructions you are viewing.
✅ Axial GRE Dixon – Correct Image Examples (4 contrasts):
Things to Look for in Axial GRE Dixon:
In-phase: overall anatomy, acquired at the echo where fat and water signals add together.
Out-of-phase: signal drop indicating fat and water in the same voxel, pointing to steatosis or a fat-containing lesion.
Water map: completely fat-suppressed, showing only water content as bright.
Fat map: shows only the signal coming from fat.
Levels match the other axial sequences, from the top of the liver dome down to the kidneys.
3D MRCP – The Cornerstone of the Study
The detailed visualization of the biliary and pancreatic duct systems helps assess ductal anatomy, stones, obstructions, congenital abnormalities, ductal dilatation, and communication with cystic lesions, all without intravenous contrast.
This sequence is normally post-processed to generate MIP (maximum intensity projection) reconstructions. The native source acquisition is what you should always review first.
✅ 3D MRCP – Correct Image Example:
Things to Look for in 3D MRCP:
The full duct tree is traceable from the intrahepatic ducts to the ampulla.
The choledochus and pancreatic duct are covered along their length.
Background signal is well suppressed, so the ducts dominate the image.
Always check the thin source images alongside the MIP. MIP reconstruction can hide small stones and exaggerate strictures.
Any apparent narrowing is cross-checked against the source images and other planes to rule out vascular pulsation from the hepatic or gastroduodenal artery.
Radial SS-TSE FS – Fast Projection Backup
The heavily T2-weighted thick slab produces projection-like images of the biliary and pancreatic ducts. It is effective for detecting biliary obstruction, choledocholithiasis, ductal dilatation, pancreatic duct abnormalities, and gallbladder pathology, with an appearance similar to conventional cholangiography in only a few seconds.
✅ Radial SS-TSE FS – Correct Image Examples:
Things to Look for in Radial SS-TSE FS:
The first slab is the central one, giving a true coronal view through the biliary tree.
Each following slab is rotated by a fixed angle, building a full 360° set of projection views.
The cross-reference on the axial plane shows which slab you are viewing.
Contrast is similar to the 3D MRCP, highlighting the same duct structures.
Detail is coarser than the 3D MRCP. That is the expected trade for motion resistance.
Final Checks:
Before finishing an MRCP, always check these 5 points to ensure diagnostic quality:
Complete Duct Coverage: The intrahepatic ducts, common hepatic duct, cystic duct, common bile duct to the ampulla, and main pancreatic duct must all be included on the 3D MRCP.
Motion Control: Respiratory-triggered sequences must show consistent triggering with no ghosting, and breath-hold sequences must show no blurring from a broken breath-hold.
Source Images Reviewed: Every finding on the MIP must be confirmed on the thin 3D source images, and every suspected stone on the SS-TSE must be confirmed on the 3D MRCP.
Consistent Planning: All axial sequences must sit at matching slice positions so contrasts can be compared level for level.
Image Quality and Artifacts: Fat suppression must be uniform, background GI fluid adequately dark, and there must be no significant chemical shift, wraparound, or susceptibility artifacts.
.svg){kind=small}
