Brain MRI Protocol: 10 Essential Steps for Sharp Screening Scans
Sequences used
Axial/sagittal T1, axial/coronal T2 FLAIR (TI ≈2500 ms at 3T), axial T2 TSE, DWI/ADC, and post-contrast T1 when indicated. Slice thickness ≤4 mm, 10% inter-slice gap.
Contrast protocol
10–15 mL (0.1 mmol/kg) GBCA at 1.5 mL/s, 100 mL saline chaser at 1.5 mL/s, ~2-minute post-injection delay before T1 acquisition.
Artefact reduction
Superior/inferior spatial saturation bands plus R/L phase-encoding direction to control CSF and carotid flow pulsation ghosting.
Primary pitfall
Confusing residual CSF/carotid pulsation artefact with true periventricular or posterior fossa pathology, in either diagnostic direction.
Table of contents
- Introduction
- Anatomy
- MR tissue relaxation values
- Scanning technique
- Contrast media protocol
- Specific absorption rate
- Top 10 pathologies
- Pitfalls — radiographers
- Pitfalls — radiologists
- Pitfalls — non-radiology physicians
- Pitfall comparison summary
- AI & automation
- Further reading
- Reducing artefacts with patients and parameters
- Parallel imaging protocols and parameters
- Conclusion
- References
Introduction
The brain MRI protocol used for routine screening is the single most frequently performed neuroimaging examination in modern radiology departments, and it is also the protocol most often taken for granted. Because it appears simple on the requisition — “MRI brain, rule out pathology” — it is easy to underestimate how many technical decisions separate a mediocre scan from a genuinely diagnostic one. Every parameter, from FLAIR inversion time to phase-encoding direction, has a direct downstream effect on whether a radiologist can confidently exclude or detect disease.
This article walks through the complete routine brain protocol as practised across 1.5T and 3.0T platforms: anatomy, tissue relaxation behaviour, ten-step scanning technique, contrast dynamics, radiofrequency safety, the ten pathologies radiographers encounter most often, and a full three-tier pitfall framework spanning scanning, interpretation, and clinical referral errors. It is written for radiographers executing the protocol daily, radiologists interpreting it, and hospital administrators overseeing departmental quality and throughput.
Routine brain MRI is requested for an enormous range of indications: chronic headache, unexplained cognitive change, seizure work-up, pre-operative planning, incidental finding follow-up, and general neurological screening. Because the differential is so broad, the routine protocol is deliberately built to be a sensitive generalist study rather than a narrow, disease-specific one — it must reliably surface white matter disease, mass lesions, vascular abnormality, and structural anomaly within a single 15–20 minute acquisition, without the luxury of the disease-specific sequences used in dedicated stroke, tumour, or epilepsy protocols.
Understanding why each parameter is set the way it is — rather than simply following a saved protocol card — is what allows a radiographer to troubleshoot confidently when a patient cannot tolerate the standard positioning, or when an unexpected finding demands an on-the-fly sequence addition. That working knowledge is the purpose of this article.
Anatomy
A routine brain MRI must adequately depict the entire supratentorial and infratentorial compartment, from the vertex to the foramen magnum. The cerebral hemispheres are divided into frontal, parietal, temporal, and occipital lobes, each with a characteristic grey matter cortical ribbon and underlying white matter tracts. The basal ganglia (caudate, putamen, globus pallidus) and thalami sit centrally around the third ventricle, while the internal capsule carries the major corticospinal and thalamocortical projections between them. Correct identification of these deep grey structures is essential, since lacunar infarcts and metabolic disease disproportionately affect this territory.
The ventricular system — lateral, third, and fourth ventricles connected by the foramina of Monro and the cerebral aqueduct — is a critical anatomical landmark because cerebrospinal fluid (CSF) flow through this system is the dominant source of the pulsation artefact this protocol is specifically tuned to control. The posterior fossa houses the brainstem (midbrain, pons, medulla) and cerebellum, structures frequently under-evaluated on suboptimal protocols due to their proximity to bone and CSF-filled spaces, and to beam-hardening-like susceptibility effects near the petrous temporal bone.
Clinically relevant vascular anatomy
The circle of Willis, formed by the anterior and posterior cerebral arteries, anterior and posterior communicating arteries, and internal carotid arteries, is the anatomical hub for most acute and chronic ischaemic pathology and the vascular structure radiographers must understand to anticipate flow artefact. The dural venous sinuses — superior sagittal, transverse, sigmoid, and straight sinus — run at the periphery of the brain and are a common site of flow-related artefact on both T2 and FLAIR. The carotid siphon, where the internal carotid artery curves through the cavernous sinus, is a well-recognised source of the pulsation ghosting this protocol’s saturation bands are designed to suppress.
Meningeal and CSF-space anatomy
The three meningeal layers — dura mater, arachnoid mater, and pia mater — define the subdural and subarachnoid spaces, both important for detecting extra-axial haemorrhage or collections. The basal cisterns (suprasellar, ambient, prepontine) are CSF-filled reservoirs at the skull base that must be carefully assessed for effacement, a subtle but critical sign of raised intracranial pressure or mass effect. Because these cisterns are small and CSF-filled, they are also disproportionately affected by residual pulsation signal when saturation bands are omitted.
Cranial nerve and skull base considerations
While a full cranial nerve evaluation belongs to dedicated protocols (internal auditory canal, orbit), the routine screen must still adequately depict the skull base, cavernous sinus, and pituitary fossa, since incidental pituitary microadenoma, vestibular schwannoma, or skull base meningioma are not infrequent incidental findings. Adequate FOV planning at the localiser stage is what prevents these structures being clipped at the inferior margin of the acquisition volume.
Standardise your neuro-anatomy positioning workflow
Consistent patient set-up reduces re-scans and protects anatomical coverage across every brain study.
Explore Positioning Solutions →MR tissue relaxation values
Understanding approximate T1 and T2 relaxation times at both field strengths is essential for predicting signal behaviour, troubleshooting unexpected contrast, and recognising when a sequence has been mis-weighted. T1 values increase with field strength while T2 values change only modestly.
| Tissue | T1 (ms) 1.5T | T1 (ms) 3.0T | T2 (ms) 1.5T | T2 (ms) 3.0T |
|---|---|---|---|---|
| Grey matter | ~1,100 | ~1,400–1,600 | ~95 | ~85–90 |
| White matter | ~750–800 | ~950–1,100 | ~75–80 | ~65–70 |
| CSF | ~4,000–4,500 | ~4,300 | ~2,000+ | ~2,000+ |
| Fat | ~250–260 | ~340–380 | ~85 | ~65–80 |
| Muscle | ~870–900 | ~1,400 | ~45–50 | ~30–35 |
| Blood (deoxygenated, venous) | ~1,200–1,400 | ~1,600 | ~150–200 (varies with oxygenation) | ~50–100 |
The FLAIR sequence’s diagnostic power comes directly from these numbers: because CSF has an extremely long T1, an inversion pulse timed at the CSF null point (TI ≈2500 ms at 3T, roughly 2000–2200 ms at 1.5T) suppresses its signal to near zero while grey and white matter, with much shorter T1 values, remain bright. This is precisely why the protocol’s critical-parameter target specifies a 3T FLAIR TI near 2500 ms — using the 1.5T inversion time on a 3T system leaves residual CSF signal that can obscure or mimic periventricular disease, one of the most consequential and preventable technical errors on this protocol.
Blood relaxation values are field- and oxygenation-dependent, which is why susceptibility-sensitive sequences (T2*, SWI) are so useful for detecting microhaemorrhage: deoxyhaemoglobin’s paramagnetic effect shortens T2* far more aggressively than it shortens T1, producing blooming artefact that is diagnostically exploited rather than simply tolerated.
Scanning technique
The following ten-step sequence reflects standard clinical workflow for a routine brain MRI, from patient set-up through to quality-control review.
- Patient screening and preparation. Confirm MRI safety screening is complete, remove all ferromagnetic items, and confirm any implanted devices are MR-conditional at the field strength being used.
- Coil selection and positioning. Position the patient supine in a dedicated head coil with the head centred at isocentre; use foam padding to minimise voluntary motion.
- Localiser/scout acquisition. Acquire a rapid three-plane localiser to confirm centring and to plan subsequent slice angulation off the anterior commissure–posterior commissure (AC-PC) line.
- Phase-encoding direction selection. Set phase-encoding to the right-left (R/L) direction for axial sequences to displace CSF and carotid pulsation ghosting away from central brain structures.
- Saturation band placement. Apply superior and inferior spatial saturation bands across the neck and vertex to suppress inflow and pulsatile arterial signal before it enters the imaging volume.
- Axial T2/FLAIR acquisition. Acquire axial T2 TSE and FLAIR with slice thickness ≤4 mm and a 10% inter-slice gap, angled parallel to the AC-PC line for reproducible anatomical comparison across studies.
- Diffusion-weighted imaging (DWI). Acquire axial DWI with b=0 and b=1000 s/mm² and generate an apparent diffusion coefficient (ADC) map to screen for acute restricted diffusion.
- Sagittal and axial T1 acquisition. Acquire pre-contrast T1-weighted sequences for baseline anatomical detail and to establish a comparison point should post-contrast imaging be added.
- Contrast decision point. If clinically indicated, proceed to the gadolinium injection protocol (see Contrast media protocol) and repeat T1 imaging post-contrast.
- Quality control review. Review all sequences at the console for motion artefact, adequate anatomical coverage from vertex to foramen magnum, and correct FLAIR CSF suppression before releasing the patient.
Each step exists to eliminate a specific, foreseeable failure mode. Step 4 and step 5 together are the direct technical answer to this protocol’s named primary artefact, while step 10’s quality-control review is what catches a mis-set TI or missed vertex coverage before the patient has left the department — far cheaper than a recall.
Scanner comparison: 1.5T vs 3.0T for routine brain MRI
| Parameter | 1.5T | 3.0T |
|---|---|---|
| FLAIR TI | ~2000–2200 ms | ~2500 ms |
| Signal-to-noise ratio (SNR) | Baseline | ~1.8–2× higher |
| Susceptibility artefact | Lower | Higher (skull base, sinuses) |
| SAR (RF heating) | Lower | Approximately 4× higher for equivalent flip angle |
| Chemical shift artefact | Lower | Doubled relative to 1.5T |
| Typical routine scan time | 15–20 minutes | 12–16 minutes (higher SNR allows faster acquisition) |
Contrast media protocol
Not every routine brain MRI requires gadolinium, but when a mass lesion, infection, demyelinating disease, or post-operative change is suspected, contrast is essential for full diagnostic yield. This protocol specifies a standard-dose injection suitable for routine screening indications.
Volume: 10–15 mL (0.1 mmol/kg body weight) gadolinium-based contrast agent.
Flow rate: 1.5 mL/s via power injector or controlled manual push.
Saline chaser: 100 mL normal saline at 1.5 mL/s to fully clear the contrast column from the injection line and peripheral vein.
Post-injection delay: approximately 2 minutes before starting post-contrast T1 acquisition, allowing adequate parenchymal and lesional enhancement to develop.
Slower flow rates such as this 1.5 mL/s target are appropriate for routine parenchymal enhancement, in contrast to angiographic or perfusion studies that demand much higher rates. A consistent, reproducible flow rate — supported by a reliable power injector and multi-use line set — reduces variability in enhancement pattern between studies on the same patient over time.[4]
Before any GBCA administration, confirm estimated glomerular filtration rate (eGFR) status, screen for prior contrast reaction, and confirm pregnancy status where relevant. Patients with an eGFR below 30 mL/min/1.73m² or acute kidney injury require risk-benefit discussion given the historical association between GBCAs and nephrogenic systemic fibrosis.[4] Macrocyclic agents are preferred where risk factors are present, consistent with international renal-safety guidance.[6]
Line hygiene and priming discipline also matter clinically, not just economically: inadequate priming or bubble management in the injector line is a recognised, preventable source of venous air embolism during contrast-enhanced imaging, reinforcing why standardised multi-use line sets and pre-injection checks belong in every department’s routine brain protocol, not only in high-flow angiographic studies.
Standardise gadolinium delivery across every scanner
Consistent injection flow rate and saline chase quality protect enhancement reproducibility scan after scan.
Explore SATJect Injection Systems →Specific absorption rate
Specific absorption rate (SAR) quantifies the radiofrequency energy deposited in tissue per unit mass, expressed in W/kg. Brain MRI protocols with multiple T2/FLAIR sequences and high refocusing flip angles can approach regulatory SAR limits, particularly at 3.0T, where energy deposition scales approximately with the square of field strength for equivalent flip angle.
| Operating mode | Whole-body SAR limit | Head SAR limit |
|---|---|---|
| Normal operating mode | 2.0 W/kg (whole body, averaged over 6 min) | 3.2 W/kg (head, averaged over 6 min) |
| First-level controlled mode | 4.0 W/kg | 3.2 W/kg |
| Second-level controlled mode | Above 4.0 W/kg — requires ethics/medical supervision | Above 3.2 W/kg |
These thresholds broadly reflect the IEC 60601-2-33 framework adopted internationally[12] and referenced in ICRP[10] and EC Radiation Protection 185[11] guidance on non-ionising radiation safety in MRI, alongside the American Association of Physicists in Medicine’s (AAPM) practice recommendations for RF safety monitoring.[13]
Five dose (SAR) reduction strategies
- Reduce refocusing flip angle on TSE/FLAIR sequences using variable flip-angle echo trains, which substantially lowers SAR with minimal image-quality trade-off.
- Increase echo spacing modestly where clinically acceptable, reducing RF duty cycle without materially degrading spatial resolution.
- Use parallel imaging acceleration to reduce the number of refocusing pulses required per slice, directly lowering total RF energy deposition.
- Select hyperecho or SPACE-type variable-flip-angle 3D sequences where volumetric imaging is required, since these inherently operate at lower average SAR than conventional 2D TSE.
- Allow adequate inter-sequence cooling time and avoid unnecessary sequence repetition, since SAR is calculated as a rolling 6-minute average — spacing acquisitions appropriately keeps cumulative deposition within normal operating mode.
Keep every brain protocol within normal SAR operating mode
Protocol-level SAR monitoring tools help radiographers stay compliant without sacrificing image quality.
Explore SAR Monitoring Resources →Top 10 pathologies
The following ten conditions represent the pathologies most frequently detected — or missed — on a routine screening brain MRI, along with their characteristic T1/T2 signal behaviour and how protocol choices affect detection.
Chronic small vessel ischaemic disease
T1: mildly hypointense. T2/FLAIR: hyperintense periventricular/deep white matter foci.
Protocol impact: requires correctly nulled FLAIR CSF signal to avoid confusing periventricular disease with residual CSF artefact.
Acute ischaemic infarct
T1: subtle hypointensity. T2/FLAIR: hyperintense; DWI: markedly restricted with low ADC.
Protocol impact: DWI/ADC inclusion is essential — FLAIR alone can be falsely normal in the hyperacute window.
Meningioma
T1: isointense to grey matter. T2: variable, often isointense; avidly enhancing post-contrast.
Protocol impact: post-contrast T1 with adequate delay is required to demonstrate the characteristic dural tail.
Glioblastoma / high-grade glioma
T1: heterogeneous, hypointense necrotic core. T2/FLAIR: hyperintense with surrounding vasogenic oedema.
Protocol impact: adequate slice coverage and post-contrast imaging needed to define irregular ring enhancement.
Metastatic disease
T1: hypo/isointense. T2/FLAIR: hyperintense lesion with disproportionate surrounding oedema.
Protocol impact: post-contrast T1 sensitivity is dose- and delay-dependent; standard 0.1 mmol/kg dosing at 2-minute delay is calibrated for this.
Demyelinating disease (e.g. multiple sclerosis)
T1: hypointense “black holes” in chronic lesions. T2/FLAIR: ovoid periventricular hyperintensities.
Protocol impact: FLAIR is the single most sensitive sequence — correct TI selection is critical for lesion conspicuity.
Cerebral microbleeds / haemorrhage
T1: variable by age of blood. T2*: marked susceptibility blooming (low signal).
Protocol impact: routine protocols benefit from a T2* or SWI add-on for full sensitivity, though not always included in the base screen.
Hydrocephalus
T1: CSF hypointense as normal. T2: ventricular enlargement, periventricular transependymal signal on FLAIR.
Protocol impact: full ventricular coverage and correct CSF nulling on FLAIR are essential to distinguish true hydrocephalus from atrophy.
Cerebral abscess
T1: hypointense centre. T2: hyperintense with a hypointense rim; DWI: markedly restricted centrally.
Protocol impact: DWI is the key differentiator from necrotic tumour, reinforcing why DWI belongs in every routine protocol.
Cerebral atrophy / neurodegenerative change
T1: useful for volumetric grey/white differentiation. T2/FLAIR: variable associated white matter change.
Protocol impact: reproducible AC-PC angulation across studies is essential for meaningful serial comparison.
Several of these entities — glioblastoma, metastatic disease, and abscess in particular — can appear radiologically similar on a single post-contrast T1 image. Differentiating them reliably depends on the full protocol working together: DWI/ADC to separate abscess (restricted) from necrotic tumour (facilitated diffusion centrally), FLAIR oedema pattern to weigh tumour grade, and precise contrast timing to characterise enhancement architecture.
Build confidence in early pathology detection
Consistent protocol execution is the foundation every downstream diagnosis depends on.
Explore Neuro-Imaging Solutions →Pitfalls — radiographers
The dominant scanning-side pitfall for the routine brain protocol is CSF and carotid flow pulsation artefact. When spatial saturation bands are omitted or phase-encoding is left in the default anterior-posterior orientation, pulsatile flow from the carotid siphon and CSF motion through the basal cisterns propagates ghosting directly across the brainstem and posterior fossa — precisely where subtle pathology is easiest to miss.
| Category | Description | Mitigation |
|---|---|---|
| Pulsation artefact | CSF/carotid flow ghosting propagates across posterior fossa and brainstem on default phase-encoding orientation. | Apply superior/inferior saturation bands; switch phase-encoding to R/L. |
| FLAIR TI mismatch | Using a 1.5T-calibrated inversion time on a 3.0T system leaves residual CSF signal. | Confirm field-strength-specific TI (~2500 ms at 3T) is loaded in the protocol card. |
| Slice angulation drift | Failure to align to the AC-PC line reduces reproducibility for serial comparison. | Use the sagittal localiser explicitly to plan AC-PC angulation every study. |
| Incomplete coverage | Vertex or posterior fossa cut off due to poor FOV planning. | Confirm full brain coverage on scout images before acquisition. |
| Motion contamination | Patient movement during FLAIR/DWI degrades diagnostic quality, especially in confused or paediatric patients. | Use immobilisation padding and consider motion-correction sequences for at-risk patients. |
| Contrast timing error | Starting post-contrast T1 before the ~2-minute delay reduces lesion conspicuity. | Standardise and time the delay explicitly rather than starting “as soon as ready.” |
Pitfalls — radiologists
The most consequential radiologist-side pitfall is mistaking residual CSF pulsation artefact for true periventricular or posterior fossa pathology — or, in the opposite direction, dismissing a genuine subtle lesion as artefact because pulsation ghosting is such a common, expected finding on brain MRI.
| Pitfall | Mechanism | Consequence | Mitigation |
|---|---|---|---|
| Artefact/pathology confusion | Ghosting from CSF pulsation overlaps anatomically with common disease sites (periventricular white matter, brainstem). | False positive or false negative reporting. | Cross-reference the DWI and T1 sequences; confirm the finding is not confined to the phase-encoding axis. |
| Satisfaction of search | Once an obvious finding (e.g. large infarct) is identified, subtler concurrent pathology is overlooked. | Missed secondary diagnosis. | Apply a structured search pattern across every sequence, every time. |
| FLAIR-only reliance in hyperacute stroke | FLAIR can remain normal in the first few hours of ischaemic stroke. | Falsely reassuring report in early presentation. | Always correlate with DWI/ADC before excluding acute infarct. |
| Enhancement timing misinterpretation | Reading post-contrast T1 acquired outside the standard delay window against normative enhancement patterns. | Under- or over-calling lesion enhancement. | Confirm the actual post-injection delay used before final interpretation. |
| Age-related change over-called as disease | Normal age-related white matter change resembles low-grade demyelinating or ischaemic disease. | Unnecessary follow-up imaging and patient anxiety. | Correlate with patient age and clinical context; use established grading scales (e.g., Fazekas) consistently. |
Pitfalls — non-radiology physicians
| Pitfall | What they see | What it actually is | Clinical danger | What to do |
|---|---|---|---|---|
| “Normal” report reassurance in acute stroke | A report stating “no acute abnormality” | May reflect a FLAIR-only read in the hyperacute window, before DWI positivity develops | Delayed stroke diagnosis and missed treatment window | Confirm DWI/ADC was specifically reviewed and correlate with clinical exam, not report wording alone |
| Assuming contrast was given | A report describing “no enhancing lesion” | Contrast may not have been administered if not clinically indicated at referral | False reassurance regarding tumour or infection | Check the technique paragraph to confirm contrast was actually administered |
| Over-reliance on white matter hyperintensity language | “Nonspecific white matter changes” in the report | Often age-appropriate small vessel change, not necessarily pathological | Unnecessary patient alarm or unwarranted specialist referral | Discuss the finding’s clinical significance directly with radiology if uncertain |
| Ignoring incidental findings | A brief mention of an incidental lesion (e.g., small meningioma) | May represent a finding requiring interval follow-up imaging | Lost to follow-up, delayed diagnosis of growth | Ensure incidental findings are documented in the patient’s active problem list with a follow-up plan |
| Assuming all brain MRIs are equivalent | A report from a routine screening protocol | Routine protocols are not optimised for every indication (e.g., epilepsy, pituitary disease) | Inadequate sensitivity for the specific clinical question | Specify the clinical question clearly on the request so the correct dedicated protocol is used |
Bridge the communication gap between radiology and referrers
Clear protocol documentation helps every clinician interpret the report correctly the first time.
Explore Referrer Education Tools →Pitfall comparison summary
🟡 Scanning (radiographers)
CSF/carotid pulsation artefact from incorrect phase-encoding or missing saturation bands is the dominant technical failure mode, compounded by FLAIR TI mismatch across field strengths.
🔴 Interpretation (radiologists)
Confusing pulsation artefact with true pathology — in either direction — alongside satisfaction-of-search errors and premature exclusion of hyperacute stroke on FLAIR alone.
🟣 Clinical (physicians)
Misreading report language, assuming contrast status without checking technique notes, and losing incidental findings to follow-up.
AI & automation
Artificial intelligence tools are increasingly embedded into the routine brain MRI workflow, particularly for stroke triage and quantitative white matter lesion tracking. A 2024 systematic review and meta-analysis of AI for stroke detection on brain MRI found that the objectives of such reviews are to estimate current detection performance for clinically representative studies and to characterise whether the underlying algorithms have received CE marking or FDA approval.[3] This reflects a broader shift: AI is no longer experimental in neuroradiology, it is a regulated, audited layer sitting alongside the radiologist.
The FDA’s review pathway for these tools has matured considerably. A 2024 Radiology commentary on FDA oversight of radiologic AI algorithms describes a real-world case in which an FDA-cleared AI algorithm misdiagnosed a finding as intracranial haemorrhage in a patient later confirmed to have ischaemic stroke[1] — a cautionary example underscoring that AI-assisted findings still require radiologist verification rather than autonomous action. A parallel 2024 review of FDA-approved stroke-triage software as medical devices reported that the great majority of the 29 devices analysed received clearance between 2018 and 2024, with adoption accelerating sharply from 2020 onward.[2]
Most currently FDA-cleared and CE-marked AI triage tools for brain imaging (e.g., large-vessel-occlusion detection, intracranial haemorrhage flagging, ASPECTS scoring) were validated primarily on CT and CTA. MRI-specific deep-learning stroke detection is an active and rapidly maturing research area, but MRI AI tools should be treated as a decision-support adjunct — never a substitute for DWI/ADC review by a qualified radiologist.
Beyond stroke triage, quantitative white matter lesion volumetry tools are being adopted for longitudinal demyelinating disease monitoring, and automated brain segmentation software increasingly supports volumetric atrophy tracking in neurodegenerative work-up. As these tools proliferate, department-level governance — documenting which tools are validated, which vendor, and which regulatory pathway — becomes a hospital administration responsibility as much as a clinical one.
Bring AI decision-support into your neuro-imaging workflow responsibly
Evidence-based triage tools support — never replace — radiologist judgement on every routine brain study.
Explore AI Neuro-Imaging Solutions →Further reading
- Gadolinium-Enhanced MRI in Brain Metastases: Enhancement Patterns, Imaging Protocols, and AI Radiomics Applications
- 7 Proven Strategies for Optimizing MRI Sequences in 2026
- 7 Critical CTA Brain & Carotids Protocol Steps Every Radiographer Must Master
- Understanding Venous Air Embolism in Contrast-Enhanced Imaging
- Top 100 Free Radiology Websites in 2026: A Global Guide for Clinicians & Radiographers
Reducing artefacts with patients and parameters
The most critical scanning parameters that impact image quality fall into four interlinked categories: spatial resolution, signal-to-noise ratio, image contrast, and artefact control. Understanding how they trade off against one another is what separates a technically competent radiographer from one who can troubleshoot a difficult brain study in real time.
Spatial resolution
Spatial resolution defines the ability to distinguish small details in an image. Matrix size (frequency × phase) increases spatial resolution as it grows, but decreases SNR because voxel size shrinks. Field of view (FOV) reduction increases spatial resolution but similarly reduces SNR through smaller voxels. Slice thickness reduction improves resolution and reduces partial volume averaging, but significantly decreases SNR — a direct trade-off against the ≤4 mm slice specification in this protocol.
Signal-to-noise ratio (SNR)
SNR represents diagnostic signal strength relative to background noise; a high SNR produces crisp images while a low SNR looks grainy. Number of averages (NEX/NSA) improves SNR by repeating acquisition, but doubling averages roughly doubles scan time. Receiver bandwidth reduction boosts SNR by limiting recorded noise, at the cost of longer scan times and increased chemical shift artefact. Coil selection — a dedicated head coil rather than a whole-body coil — captures substantially stronger signal and heavily improves SNR.
Image contrast
Repetition time (TR) is the interval between RF pulses; a short TR maximises T1 contrast while a long TR minimises it. Echo time (TE) is the interval between the RF pulse and echo peak; a short TE minimises T2 effects while a long TE maximises T2 weighting, making CSF and fluid appear bright — exactly the behaviour exploited in the FLAIR sequence’s suppression logic. Flip angle controls proton excitation and is especially critical in gradient echo sequences.
Artefact control
Phase-encoding direction swapping shifts motion-induced artefacts, such as CSF pulsation, away from the region of interest — the exact remedy specified for this protocol’s primary artefact. Flow compensation and gating use physiological triggers to minimise blurring from pulsatile motion. Parallel imaging uses multiple coil elements simultaneously to reduce phase-encoding steps, cutting scan time and reducing motion artefact susceptibility.
Practical patient-side technique
Beyond scanner parameters, patient-side technique meaningfully reduces artefact burden. Clear pre-scan communication about breath-holding is not relevant to brain imaging, but swallowing and head-nodding instructions are: asking the patient to remain still and avoid swallowing during FLAIR and DWI acquisition measurably reduces motion-related blurring. Adequate padding at the temples and under the knees reduces the residual postural fidgeting that commonly compromises the final two minutes of a longer sequence.
Parallel imaging protocols and parameters
Parallel imaging acceleration (GRAPPA, SENSE, ASSET, and vendor equivalents) reduces the number of phase-encoding steps acquired, shortening scan time and reducing susceptibility to motion — but at a direct SNR cost that scales with the acceleration (turbo/GRAPPA) factor.[25] Selecting the correct factor for each sequence and field strength is a balancing act between speed, SNR, and residual aliasing artefact.
| Sequence | 1.5T typical GRAPPA/turbo factor | 1.5T adjustment needed | 3.0T typical GRAPPA/turbo factor | 3.0T adjustment needed |
|---|---|---|---|---|
| Axial FLAIR | Turbo factor 15–19; GRAPPA 2 | Increase NEX slightly to offset SNR loss | Turbo factor 15–19; GRAPPA 2–3 | 3T’s inherent SNR surplus absorbs the acceleration penalty better than 1.5T |
| Axial T2 TSE | Turbo factor 12–16; GRAPPA 2 | Modest bandwidth increase reduces chemical shift | Turbo factor 12–16; GRAPPA 2 | Reduce refocusing flip angle to manage SAR at higher acceleration |
| DWI (EPI-based) | GRAPPA 2 | Keep acceleration conservative to limit distortion | GRAPPA 2–3 | Higher GRAPPA factor helps offset increased susceptibility distortion at 3T |
| Post-contrast T1 | GRAPPA 2 | Standard acceleration; minimal trade-off | GRAPPA 2–3 | Use to shorten breath-independent T1 acquisition without SNR penalty given 3T baseline signal |
As a rule of thumb, higher turbo/GRAPPA factors are better tolerated at 3.0T because the field strength’s inherent SNR surplus offsets the acceleration-related signal loss, whereas at 1.5T aggressive acceleration factors risk pushing SNR below the diagnostic threshold, particularly on FLAIR and DWI where noise directly degrades lesion conspicuity.
Conclusion
The routine brain MRI protocol is deceptively simple in name but technically demanding in execution. Correct FLAIR inversion timing calibrated to field strength, disciplined slice angulation along the AC-PC line, and deliberate phase-encoding and saturation-band strategy to control CSF and carotid pulsation artefact together determine whether the study reliably captures the ten pathologies covered in this article — from acute infarct to chronic small vessel disease to incidental mass lesions.
The three-tier pitfall framework presented here — scanning-side pulsation artefact, interpretation-side artefact/pathology confusion, and clinical-side report misreading — reflects the reality that diagnostic quality is a shared responsibility across the entire imaging chain, not a single technologist’s or radiologist’s task alone. A protocol executed with this level of discipline gives every downstream clinician the confidence that a “normal” report truly means normal.
References
- Zhang, K., Khosravi, B., Vahdati, S., & Erickson, B. J. (2024). FDA review of radiologic AI algorithms: Process and challenges. Radiology, 310(1), e230242. https://doi.org/10.1148/radiol.230242
- Haq, I. U., et al. (2024). Revolutionizing acute stroke care: A review of Food and Drug Administration-approved software as medical devices for stroke triage. Cureus, 16(11), e74686. https://doi.org/10.7759/cureus.74686
- Bojsen, J. A., Elhakim, M. T., & Graumann, O. (2024). Artificial intelligence for MRI stroke detection: A systematic review and meta-analysis. Insights into Imaging, 15, 160. https://doi.org/10.1186/s13244-024-01723-7
- Joint Committee for NSF and Use of Gadolinium-Based Contrast Agents. (2025). Guidelines for administering gadolinium-based contrast agents to patients with renal dysfunction (Version 3, revised 2024). Japanese Journal of Radiology. https://doi.org/10.1007/s11604-024-01719-9
- Authors withheld for peer review. (2025). From routine to selective: How updated MRI guidelines reshape gadolinium use in Germany. PMC. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12087169/
- Authors withheld for peer review. (2026). Safety of a tailored gadolinium-based contrast agent protocol considering excretion pathways in patients with renal impairment. Diagnostics, 16(3), 451. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12897084/
- Almudayni, A., Alharbi, M., Chowdhury, A., Ince, J., Alablani, F., Minhas, J. S., Lecchini-Visintini, A., & Chung, E. M. L. (2022). Magnetic resonance imaging of the pulsing brain: A systematic review. Magnetic Resonance Materials in Physics, Biology and Medicine, 35, 651–668. https://doi.org/10.1007/s10334-022-01043-1
- American College of Radiology. (2023). ACR Manual on Contrast Media. https://www.acr.org/Clinical-Resources/Contrast-Manual
- American College of Radiology, American Society of Neuroradiology, & Society for Pediatric Radiology. (2021). ACR–ASNR–SPR practice parameter for the performance of magnetic resonance imaging of the adult brain. https://www.acr.org/-/media/ACR/Files/Practice-Parameters/mr-brain.pdf
- International Commission on Radiological Protection. (2021). ICRP guidance on non-ionising radiation and MRI safety. https://www.icrp.org/
- European Commission, Directorate-General for Energy. (2018). Radiation Protection No. 185: Guidelines for MRI safety and appropriate use. Publications Office of the European Union. https://energy.ec.europa.eu/publications/radiation-protection-185_en
- International Electrotechnical Commission. (2015). IEC 60601-2-33: Particular requirements for the basic safety and essential performance of magnetic resonance equipment for medical diagnosis. https://webstore.iec.ch/publication/22705
- American Association of Physicists in Medicine. (2020). AAPM report: MR safety practice guidelines. https://www.aapm.org/pubs/reports/
- European Society of Radiology. (2022). ESR statement on the validation of imaging biomarkers. Insights into Imaging, 13, 137. https://doi.org/10.1186/s13244-022-01279-4
- Radiological Society of North America. (2021). RSNA Quantitative Imaging Biomarkers Alliance (QIBA) MRI profile documentation. https://www.rsna.org/research/quantitative-imaging-biomarkers-alliance
- American Heart Association / American Stroke Association. (2019). Guidelines for the early management of patients with acute ischemic stroke. Stroke, 50(12), e344–e418. https://doi.org/10.1161/STR.0000000000000211
- European Stroke Organisation. (2021). ESO guideline for the management of extracranial and intracranial artery dissection. European Stroke Journal, 6(3), XXXIX–LXXXVIII. https://doi.org/10.1177/23969873211046475
- Filippi, M., Preziosa, P., Banwell, B. L., et al. (2019). Assessment of lesions on magnetic resonance imaging in multiple sclerosis: Practical guidelines. Brain, 142(7), 1858–1875. https://doi.org/10.1093/brain/awz144
- Wattjes, M. P., Ciccarelli, O., Reich, D. S., et al. (2021). 2021 MAGNIMS-CMSC-NAIMS consensus recommendations on the use of MRI in patients with multiple sclerosis. The Lancet Neurology, 20(8), 653–670. https://doi.org/10.1016/S1474-4422(21)00095-8
- Vernooij, M. W., Ikram, M. A., Vrooman, H. A., et al. (2015). Incidental findings on brain MRI in the general population: Long-term follow-up and clinical relevance. Radiology, 275(3), 720–730. https://doi.org/10.1148/radiol.2015140970
- Wardlaw, J. M., Smith, E. E., Biessels, G. J., et al. (2019). Neuroimaging standards for research into small vessel disease, update. The Lancet Neurology, 18(7), 684–696. https://doi.org/10.1016/S1474-4422(19)30079-1
- Runge, V. M. (2017). Dechelation and residual gadolinium: Practical implications for radiologists and referring clinicians. Investigative Radiology, 52(6), 317–323. https://doi.org/10.1097/RLI.0000000000000371
- Gulani, V., Calamante, F., Shellock, F. G., Kanal, E., & Reeder, S. B. (2017). Gadolinium deposition in the brain: Summary of evidence and recommendations. The Lancet Neurology, 16(7), 564–570. https://doi.org/10.1016/S1474-4422(17)30158-8
- Zaitsev, M., Maclaren, J., & Herbst, M. (2015). Motion artifacts in MRI: A complex problem with many partial solutions. Journal of Magnetic Resonance Imaging, 42(4), 887–901. https://doi.org/10.1002/jmri.24850
- Deshmane, A., Gulani, V., Griswold, M. A., & Seiberlich, N. (2016). Parallel MR imaging. Journal of Magnetic Resonance Imaging, 36(1), 55–72. https://doi.org/10.1002/jmri.23639
- Ebrahimian, S., Kalra, M. K., Agarwal, S., et al. (2022). FDA-regulated AI algorithms: Trends, strengths, and gaps of validation studies. Academic Radiology, 29(4), 559–566. https://doi.org/10.1016/j.acra.2021.09.002
- Kanal, E., Barkovich, A. J., Bell, C., et al. (2020). ACR guidance document on MR safe practices: Updated 2020. Journal of Magnetic Resonance Imaging, 51(5), 1287–1301. https://doi.org/10.1002/jmri.27128
