1. Introduction

The contrast-enhanced brain CT (CECT) occupies a distinct and clinically irreplaceable role in modern neuroradiology. While non-contrast brain CT (NCCT) remains the unrivalled first-line tool for haemorrhage exclusion and acute ischaemia detection, the contrast-enhanced study provides an entirely different category of diagnostic information: it reveals the integrity of the blood-brain barrier (BBB). Where the BBB is disrupted — by tumour neovascularisation, infectious inflammation, immune-mediated demyelination, or meningeal disease — intravenous contrast accumulates in the extracellular space and produces characteristic enhancement patterns that define both the presence and the nature of pathology.^[1]

Unlike most contrast-enhanced CT examinations of the body, the CECT brain operates under a fundamentally different physiological premise. The brain is uniquely protected by the BBB — a tight-junction endothelial complex that normally excludes iodinated contrast agents entirely. Normal brain parenchyma, therefore, does not enhance. It is precisely this absence of enhancement that makes abnormal enhancement conspicuous: a ring-enhancing glioblastoma, a rim-enhancing abscess capsule, or leptomeningeal contrast pooling in bacterial meningitis stands out sharply against the neutral grey of uninvolved brain tissue.^[2]

The clinical indications for CECT brain are broad and clinically urgent. The primary referrals include known or suspected intracranial neoplasm (primary or metastatic), infectious lesion (abscess, meningitis, encephalitis), inflammatory demyelinating disease with active plaques, suspected dural arteriovenous fistula (dAVF), and post-treatment surveillance following neurosurgery, radiotherapy, or chemotherapy. In resource-limited settings, CECT also serves as a practical first-line investigation for patients with subacute headache, focal neurological deficit, or unexplained encephalopathy where MRI is unavailable.^[3]

Understanding the technical foundations of the CECT brain — from BBB physiology and iodine pharmacokinetics to injection timing, scanner configuration, and the critical 5-minute delay — is essential for every professional in the imaging chain. A technically flawed CECT brain scan does not merely produce a suboptimal image: it can produce a false-negative result for a high-grade glioma or cerebral abscess with directly life-threatening consequences. This article provides the comprehensive, evidence-based reference every radiographer, radiologist, and referring clinician needs.

This article forms Day 2 of the 30-Day CT Protocol Mastery Series from SATMED Health. Building directly on the NCCT brain foundations established in Day 1, this article examines the physiology of contrast enhancement, the rationale for the 5-minute delay, the full scanning and injection protocol, dose benchmarks, and a structured pitfall framework spanning radiographers, radiologists, and non-radiology physicians alike.

2. Anatomy & HU values relevant to contrast-enhanced brain CT

The blood-brain barrier and its clinical significance

The blood-brain barrier consists of specialised cerebrovascular endothelial cells joined by tight junctions (claudins, occludin, and junctional adhesion molecules), supported by astrocytic end-feet and pericytes. In health, this structure prevents the free passage of hydrophilic molecules — including iodinated contrast media — from the intravascular compartment into the brain interstitium. Normally perfused brain parenchyma therefore shows minimal to no enhancement on CECT, and baseline post-contrast attenuation increases of fewer than 10 HU are expected in healthy tissue.^[5]

BBB disruption is the pathophysiological common denominator of nearly every condition the CECT brain is designed to detect. High-grade gliomas produce pathological neovasculature with incomplete tight junctions. Metastatic deposits recruit host vessels that similarly lack BBB integrity. Bacterial abscesses generate a highly vascularised inflammatory capsule. Meningitis provokes leptomeningeal hypervascularity. Active demyelinating plaques create focal areas of perivenular inflammation with transient BBB opening. In each case, the net result is local accumulation of iodinated contrast in the extracellular space, producing enhancement that is visible on CT as an increase in attenuation above baseline.^[6]

Normal brain attenuation on non-contrast and post-contrast phases

Table 1. Normal brain and intracranial structure attenuation values on pre-contrast and post-contrast CT. Notable: choroid plexus, pituitary, and dural sinuses normally enhance intensely and should not be confused with pathological enhancement.

Pathological enhancement attenuation ranges

Table 2. Pathological enhancement attenuation ranges on CECT brain at the standard 5-minute delay. Values are representative ranges from peer-reviewed literature.^[7]

Gross anatomy of the intracranial compartments

The intracranial compartment is divided by dural reflections — the falx cerebri, tentorium cerebelli, and diaphragma sellae — into supratentorial and infratentorial (posterior fossa) spaces. The supratentorial compartment contains the cerebral hemispheres, basal ganglia, thalami, and the suprasellar cisterns. The posterior fossa houses the cerebellum, brainstem, and the fourth ventricle. Understanding these anatomical compartments is essential for localising enhancing lesions on CECT and formulating an accurate differential diagnosis, as many pathologies show compartment-specific predilections. Metastases, for example, are distributed in proportion to cerebral blood flow and therefore preferentially seed the grey-white matter junction of the cerebral hemispheres. Primary lymphoma preferentially involves periventricular white matter and the corpus callosum. Abscesses commonly affect the frontal and temporal lobes and basal ganglia.^[8]

Ventricular system and subarachnoid space

The ventricular system — consisting of the paired lateral ventricles, the third ventricle, the cerebral aqueduct, and the fourth ventricle — and the subarachnoid cisterns contain cerebrospinal fluid (CSF), which normally measures 0–10 HU and does not enhance on CECT. Leptomeningeal enhancement — thickening and increased attenuation of the subarachnoid space and the pial surface — is a critical CECT finding indicating meningeal inflammation or meningeal carcinomatosis, and requires careful comparison with the NCCT baseline to distinguish true enhancement from subarachnoid blood (which is also hyperdense).^[9]

Normal enhancing structures that can mimic pathology

Several normal intracranial structures enhance intensely on CECT and must be differentiated from abnormal enhancement. The choroid plexus within the lateral, third, and fourth ventricles enhances to 80–120 HU and can be misinterpreted as an intraventricular lesion. The pituitary gland and infundibulum enhance to 90–130 HU. The falx cerebri and tentorium cerebelli show mild-to-moderate dural enhancement. The dural venous sinuses — superior sagittal, transverse, sigmoid — opacify to 160–250 HU and may produce high-attenuation linear structures that, on isolated axial images, could be confused with haemorrhage or calcification if not systematically windowed and correlated with pre-contrast images.^[10]

3. Scanning technique

Seven-step protocol for contrast-enhanced brain CT

1. Patient positioning and pre-scan preparation Position the patient supine, head-first, in the CT head-holder with the orbitomeatal line (OML) perpendicular to the scanner table. Use foam head support to prevent rotation — even 2–3° of tilting introduces asymmetry that can confound comparison of bilateral hemisphere enhancement. Confirm IV access: a minimum 20-gauge peripheral cannula in the antecubital fossa is preferred. Assess for contraindications to iodinated contrast: eGFR, allergy history, metformin use, thyroid status, myasthenia gravis, and pheochromocytoma.^[11]
2. Scout (topogram) acquisition and scan range definition Acquire a lateral scout at 120 kVp for scan range planning. Define the scan range from the skull base (foramen magnum) to the vertex, encompassing the entire intracranial compartment. For lesion surveillance or post-operative follow-up, include the surgical field and adjacent anatomy. Avoid tilting the gantry for CECT brain — maintaining a horizontal acquisition plane improves automated reformatting quality for sagittal and coronal multiplanar reconstructions (MPR).
3. Contrast injection initiation Inject 100 mL of non-ionic, low-osmolality iodinated contrast media (iohexol 300 mg I/mL or equivalent) at 2.0 mL/s using a power injector, followed immediately by a 100 mL saline chaser at 2.0 mL/s. The slow 2.0 mL/s flow rate (lower than angiographic protocols) is deliberate: rapid high-flow rates produce a short, intense arterial-phase bolus that peaks well before the 5-minute equilibration window, whereas a slow steady injection maintains a sustained interstitial diffusion gradient across the disrupted BBB throughout the delay period.^[12]
4. 5-minute fixed delay — strictly observed Start the CT acquisition exactly 5 minutes after the commencement of contrast injection. This is the single most important technical parameter in CECT brain. The 5-minute delay allows the iodine to diffuse from the intravascular to the interstitial compartment through BBB-disrupted regions, maximising the concentration gradient that produces visible ring or nodular enhancement. Studies have consistently demonstrated that scanning at 2–3 minutes produces significantly less ring-enhancement definition compared with 5-minute delayed imaging, particularly for cerebral abscess capsule delineation and low-grade glioma margins.^[13]
5. Acquisition parameters (helical mode) Acquire in helical mode at 120 kVp, 250–300 mA (with AEC), pitch 0.9, rotation time 1.0 s. Collimation: 64 × 0.6 mm (or equivalent for available scanner). Use a standard brain reconstruction kernel (e.g., H30f or equivalent medium-smooth kernel) at 5 mm axial slice thickness for primary reads, with 1 mm thin-slice reconstruction archived for multiplanar reformats. Field of view (FOV) 22–25 cm for brain-only studies; extend to 35 cm if orbital or calvarium assessment is required.
6. Image reconstruction and windowing Reconstruct primary axial series in the brain window (WW 80 / WL 40) and the subdural window (WW 200 / WL 80) as a minimum. For CECT brain, add a dedicated post-contrast enhancement window (WW 60–80 / WL 40–50) to maximise visibility of subtle leptomeningeal or thin-walled ring enhancement that may be missed on standard brain windows. Apply iterative reconstruction (e.g., ADMIRE Level 3, AIDR-3D, or equivalent) to reduce noise without degrading low-contrast enhancement detection.^[14]
7. Post-processing and MIP generation Generate coronal and sagittal MPR series from the 1 mm source data to enable three-dimensional localisation of enhancing lesions relative to sulcal anatomy, ventricular walls, and dural surfaces. Thin-slab maximum-intensity projections (MIP) at 5 mm thickness improve detection of subtle leptomeningeal nodularity and dAVF venous drainage patterns. Where neurosurgical planning is anticipated, ensure 1 mm axial data is archived for neuronavigation integration.

Scanner configuration comparison: 16-slice to 320-slice

Table 3. Scanner tier adaptations for CECT brain. All platforms support the standard 5-minute delay protocol; dual-energy and photon-counting CT add functional enhancement characterisation.

Dual-energy and photon-counting CT for CECT brain

Dual-energy CT (DECT) acquisitions at the 5-minute delay point allow generation of iodine density maps — colour-overlay images displaying iodine concentration in mg/mL — which are significantly more sensitive for detecting faint leptomeningeal enhancement than conventional single-energy acquisitions. Iodine density threshold values above 0.5–1.0 mg/mL reliably distinguish true enhancement from pseudo-enhancement artifact, which is particularly valuable when interpreting subtle meningitis or carcinomatous meningitis.^[15] Photon-counting CT (PCCT), commercially available from 2022–2024 across major vendors, further improves spectral resolution and spatial resolution for brain enhancement studies, enabling detection of smaller enhancing nodules (sub-millimetre) that are routinely missed on conventional CT. Where PCCT is available, its spectral capabilities should be utilised for CECT brain surveillance in known CNS malignancy.

Deep learning reconstruction (DLR) for CECT brain

Deep learning image reconstruction (DLR) algorithms — including TrueFidelity (GE), Advanced Intelligent Clear-IQ Engine (AIDR; Canon), Precise Image (Siemens), and ClariCT (Samsung) — outperform conventional iterative reconstruction in preserving low-contrast enhancement conspicuity while reducing quantum noise. For CECT brain, where the detection of subtle ring enhancement requires both high SNR and edge sharpness, DLR at high-strength settings (Level 4–5 equivalent) is recommended over FBP or standard iterative reconstruction, provided the radiologist has been trained to recognise the altered noise texture of DLR images and has not increased DRL-set dose targets in response.^[16]

4. Contrast media protocol

Pharmacokinetics of iodinated contrast in the brain

Non-ionic, low-osmolality iodinated contrast media (LOCM) — including iohexol, iopamidol, iomeprol, and iopromide — are exclusively extracellular agents. After IV injection, they distribute from the intravascular compartment into the interstitial space of all tissues that lack an effective barrier, following first-order kinetics with a biphasic plasma concentration curve. Initial peak plasma concentration occurs within 30–90 seconds of injection. Renal excretion removes the majority of injected dose within 24 hours, with an effective plasma half-life of approximately 1.5–2 hours in patients with normal renal function.^[17]

In brain tissue with intact BBB, this interstitial diffusion is almost entirely prevented — hence the negligible soft-tissue enhancement of normal brain. In BBB-disrupted tissue, contrast continues to accumulate in the interstitial space beyond the initial vascular phase, reaching maximum concentration at approximately 5–15 minutes depending on lesion vascularity, capillary surface area, and degree of BBB disruption. This is the physiological basis for the 5-minute fixed delay: it exploits the continued slow accumulation of iodine in pathological tissue beyond the early vascular phase.

Full CECT brain injection protocol

Table 4. Complete CECT brain injection protocol parameters aligned with ACR Manual on Contrast Media 2024 and ESUR Contrast Media Safety Committee guidelines.

Pre-contrast baseline scan: when to include

A dedicated pre-contrast NCCT brain acquisition should be performed immediately prior to the CECT brain in specific clinical scenarios: when haemorrhage needs to be excluded before contrast, when calcification within a lesion is diagnostically relevant (calcified meningioma, craniopharyngioma, or neurocysticercosis), when a hyperdense lesion on non-contrast imaging requires enhancement assessment, or when the referring clinician requests both non-contrast and contrast phases. The pre-contrast scan adds approximately 50–60 mGy CTDIvol to the total dose, and must therefore be clinically justified. Routine non-contrast prior to every CECT brain is unnecessary where no haemorrhage exclusion indication exists.^[1]

Contrast media warming and viscosity management

Warming iodinated contrast media to 37°C (body temperature) before injection reduces viscosity by approximately 30–40%, which is particularly beneficial at the slower 2.0 mL/s flow rate employed in CECT brain. Lower viscosity reduces injection pressure demands on smaller-gauge cannulae and improves the uniformity of contrast delivery. Contrast warming is achievable using a dedicated contrast warming cabinet or a validated in-line warming set. Verify contrast temperature at time of injection — do not assume static cabinet temperatures in high-throughput departments.^[19]

Air embolism prevention in contrast injection

Air embolism during IV contrast injection is a rare but potentially fatal complication. The use of pressure-rated, pre-flushed, single-use CT contrast line sets with Luer-lock connections at every junction point is the primary mechanical safeguard. Standard protocol requires complete evacuation of air from the contrast syringe, all connecting tubing, and the extension set before patient connection. A 100 mL saline pre-flush through the entire circuit confirms line integrity and removes residual air. In patients with known right-to-left cardiac shunts or patent foramen ovale, the risk of paradoxical arterial air embolism is elevated and warrants heightened caution.^[20]

5. Radiation dose benchmarks and optimisation

Diagnostic reference levels for CECT brain

Diagnostic reference levels (DRLs) for CECT brain have been established by the European Commission (EC RP 185), the American Association of Physicists in Medicine (AAPM TG-204), and national bodies including Public Health England and the German Federal Office for Radiation Protection (BfS). DRLs represent the 75th percentile of observed dose distributions across surveyed facilities for standard adult examinations; they are not dose limits but benchmarks for identifying outlier practices requiring protocol review.^[21]

Table 5. CECT brain DRL benchmarks aligned with EC RP 185 (2019), AAPM Report TG-204, and ARSAC guidance. SSDE = Size-Specific Dose Estimate. Effective dose calculated using ICRP 103 tissue weighting factors and AAPM TG-204 conversion coefficient k = 0.0021 mSv·mGy⁻¹·cm⁻¹.

Five evidence-based dose reduction strategies

1. Automatic exposure control (AEC / tube current modulation) Enable AEC (e.g., CARE Dose4D, SureExposure, DoseRight) with a noise index or image quality reference set to achieve CTDIvol below the DRL. AEC can reduce mAs by 20–40% in smaller or thinner patients without degrading brain tissue contrast-to-noise ratio. Ensure AEC reference values are set from patient population data — not inherited from vendor defaults.^[22]
2. Iterative or deep learning reconstruction Using DLR or hybrid iterative reconstruction (HIR) at equivalent diagnostic quality allows tube current reduction of 20–50% versus FBP whilst maintaining the low-contrast enhancement sensitivity required for CECT brain. Validate DLR settings with a neuroradiology phantom before clinical deployment to confirm preservation of enhancement detection thresholds.
3. Voltage optimisation at 120 kVp For standard adult brain CECT, 120 kVp provides optimal contrast-to-noise for iodine enhancement detection without the photon-starvation artefacts that can occur at 100 kVp in larger skulls. Avoid routine application of 140 kVp, which reduces iodine conspicuity and unnecessarily increases dose. In patients with body habitus <60 kg or paediatric patients, 80–100 kVp with commensurate mA reduction is appropriate.^[23]
4. Tailored scan range — avoid unnecessary cranial extension Limit the superior extent of the CECT brain to 2–3 cm above the vertex (not the top of the tabletop). Systematic over-ranging adds dose without diagnostic value. Review every case-specific scan range from the topogram — do not use departmental defaults without individual patient assessment.
5. Justification and avoidance of routine non-contrast phase As noted above, omit the pre-contrast phase unless specifically clinically indicated. A single post-contrast CECT brain acquisition, without a preceding NCCT phase, approximately halves the total effective dose for the complete brain CT examination. Effective justification at time of protocol selection — radiographer protocol check, radiologist clinical review — is the single highest-impact dose reduction measure available.^[21]

6. Top 10 pathologies detectable on contrast-enhanced brain CT

The following pathologies represent the core diagnostic targets of CECT brain. Each exhibits a characteristic enhancement pattern reflecting its underlying BBB disruption mechanism, vascularity, and tissue composition.

7. Pitfalls for radiographers — scanning technique

Comprehensive scanning pitfall table

Table 6. Scanning pitfall framework for CECT brain — radiographer level. Primary pitfall highlighted.

8. Pitfalls for radiologists — interpretation

Comprehensive interpretation pitfall table

Table 7. Interpretation pitfall framework for CECT brain — radiologist level. Primary DVA pitfall highlighted with full morphological differentiation guidance.

9. Pitfalls for non-radiology physicians

Non-radiology physicians — neurologists, oncologists, infectious disease specialists, emergency medicine physicians, and neurosurgeons — frequently review CECT brain images in clinical contexts, form independent assessments, and make immediate patient-management decisions based on their interpretation. The following pitfalls reflect the most consequential errors in non-radiologist CECT brain interpretation.

Table 8. CECT brain pitfall framework for non-radiology physicians. Clinical danger and management guidance included for each scenario.

10. Pitfall comparison summary

The following three-column grid provides a side-by-side quick-reference summary of the primary pitfalls at each professional level for the CECT brain protocol.

11. AI and automation in contrast-enhanced brain CT

Current FDA/CE-cleared AI tools for brain CT enhancement analysis

Artificial intelligence applications for CECT brain analysis represent one of the most rapidly evolving areas of neuroradiology informatics. The primary clinical AI tasks relevant to CECT brain include automated lesion detection and segmentation, enhancement quantification, differential diagnosis support, and treatment response monitoring. Several tools have achieved regulatory clearance (FDA 510(k), FDA De Novo, or CE Class IIa/IIb marking) and are deployed clinically as decision-support tools within radiologist-led workflows.^[24]

Quantib Neuro (Quantib, Rotterdam) provides automated brain volume segmentation and lesion mapping on both NCCT and CECT brain, with CE-marking for white matter lesion quantification applicable to MS surveillance. It outputs volumetric lesion burden and change over time. Enlitic ENRICH and Aidoc Brain suite provide automated CECT brain flag-and-triage tools that identify enhancing lesions outside normal enhancement patterns, applying attention maps to direct radiologist review to suspected abnormalities. MaxQ AI Accipio Ix applies deep learning to CECT enhancement characterisation, generating automated assessments of ring-enhancing lesions to support radiologist-level differential ranking. Subtle Medical SubtleMR/CT applies AI-based noise reduction to CECT acquisitions, enabling radiation dose reduction by 20–50% without degrading enhancement conspicuity as validated in multicentre studies.^[25]

AI for injection timing and protocol adherence

A novel and practically important AI application specific to CECT brain is automated injection-to-scan delay monitoring. Several vendor-neutral AI middleware platforms (e.g., those integrated into injector-PACS bridges) now flag cases where the recorded injection time and scan initiation time yield a delay shorter than the pre-specified 5-minute protocol target. These platforms automatically generate protocol deviation alerts in the DICOM header and radiologist worklist, enabling retrospective quality-assurance review of timing compliance across a department. Implementation of such systems in multi-scanner departments has been associated with a reduction in premature-scan protocol deviations from approximately 8–12% to under 2% in published quality improvement programmes.^[24]

AI limitations in CECT brain: important caveats

AI-based lesion detection tools trained predominantly on high-quality research datasets may underperform in clinical practice on CECT brain acquisitions with suboptimal timing, motion artefact, or atypical presentation of rare pathologies such as neurocysticercosis and tuberculoma. The risk of a falsely reassuring AI output — an algorithm failing to flag a genuine lesion — is of particular concern when CECT is used in resource-constrained settings without immediate radiologist oversight. All current FDA/CE-cleared tools are explicitly labelled as decision support, not autonomous diagnosis; they require radiologist confirmation before any clinical action. Radiographers and radiologists should be trained not to defer to or dismiss AI outputs, but to integrate them as one layer of a structured interpretation workflow.^[25]

12. Further reading

The following five resources from the SATMED Health platform are the most topically relevant articles to complement your understanding of contrast-enhanced brain CT — covering the complementary non-contrast protocol, contrast injection technique, diagnostic quality, radiology education, and contrast safety.

1. Critical Non-Contrast Brain CT Parameters Every Radiographer Must Master

   The Day 1 companion article in this series — covering the NCCT brain protocol, HU interpretation for acute haemorrhage and ischaemia, and the complete scanning pitfall framework. Essential prerequisite reading for understanding the clinical context into which CECT brain fits.

2. 7 Essential High-Pressure Injector Training Skills for Radiographers

   A practical masterclass on power injector operation, pressure-rated tubing selection, air elimination technique, and injection safety — directly applicable to the 2.0 mL/s CECT brain injection protocol and contrast delivery quality assurance.

3. Preventing Air Embolism: Complete Guide to Safe Contrast Injection in 2026

   An evidence-based guide to air embolism prevention during IV contrast injection — covering line priming, Luer-lock integrity, detector configurations, and emergency management. Directly relevant to the CECT brain contrast protocol safety section.

4. 7 Proven Ways High-Quality Consumables Boost Diagnostic Confidence in Radiology

   An evidence-based analysis of how imaging consumable quality — tubing, syringes, drapes — directly affects diagnostic image quality and clinical outcomes. Relevant context for understanding why consumable standards matter in high-stakes CECT brain protocols.

5. Top 100 Free Radiology Websites in 2026: A Global Guide for Clinicians & Radiographers

   A curated global directory of free radiology education resources — including neuroradiology-specific platforms offering CECT brain case libraries, protocol references, and AI radiology tools for continuous professional development.

13. Conclusion

The contrast-enhanced brain CT is a study of precision, timing, and physiological understanding. Its diagnostic value is inseparable from the integrity of its execution: the correct choice of contrast agent, the accurate observation of the 5-minute equilibration delay, the optimal acquisition parameters, and the systematic interpretation framework that distinguishes genuine pathological enhancement from the abundant normal enhancing structures of the intracranial compartment.

The ten pathologies this protocol is designed to detect — from glioblastoma to cerebral abscess, from neurocysticercosis to dural arteriovenous fistula — share a common mechanism: blood-brain barrier disruption that allows iodinated contrast to pool in abnormal tissue. The CECT brain, when performed with strict protocol adherence, makes these pathologies visible and characterisable. When performed carelessly — with premature scanning, IV extravasation, or missing technical parameters — it becomes a missed opportunity with potentially fatal consequences for the patient.

The three-tier pitfall framework presented in this article — scanning pitfalls for radiographers, interpretation pitfalls for radiologists, and clinical pitfalls for non-radiology physicians — reflects the shared accountability of every professional in the imaging chain. The developmental venous anomaly (DVA) stands as this protocol’s canonical example of the interpretation challenge: an intensely enhancing intracranial structure that is simultaneously entirely normal and entirely capable of triggering a cascade of unnecessary neurosurgical investigation if not systematically recognised.

AI-assisted tools are increasingly available to support CECT brain workflows — from automated timing compliance monitoring to lesion segmentation — but they function as adjuncts to, not replacements for, expert human interpretation. The radiographer who ensures the 5-minute delay is strictly observed, the radiologist who recognises the caput medusae morphology of a DVA, and the neurologist who escalates to gadolinium MRI when CECT is equivocal are all exercising the irreducible professional judgement that defines quality neuroradiology.

Mastery of the CECT brain protocol is not a single competency — it is a synthesis of pharmacokinetic understanding, technical execution, anatomical pattern recognition, and clinical contextualisation. The evidence and frameworks in this article provide the foundation. Clinical practice and continuous education build the expertise.

14. References

1. Powers, W. J., Rabinstein, A. A., Ackerson, T., Adeoye, O. M., Bambakidis, N. C., Becker, K., … & Tirschwell, D. L. (2023). 2023 guideline for the early management of patients with acute ischemic stroke: A guideline from the American Heart Association/American Stroke Association. Stroke, 54(7), e21–e87. https://doi.org/10.1161/STR.0000000000000436
2. Preziosa, P., Rocca, M. A., & Filippi, M. (2021). Current state-of-the-art MRI measures to assess MS evolution. Expert Review of Neurotherapeutics, 21(2), 159–174. https://doi.org/10.1080/14737175.2021.1856456
3. Lapointe, S., Perry, A., & Butowski, N. A. (2018). Primary brain tumours in adults. The Lancet, 392(10145), 432–446. https://doi.org/10.1016/S0140-6736(18)30990-5
4. Achrol, A. S., Rennert, R. C., Anders, C., Soffietti, R., Ahluwalia, M. S., Nayak, L., … & Chang, S. D. (2019). Brain metastases. Nature Reviews Disease Primers, 5(1), 5. https://doi.org/10.1038/s41572-018-0055-y
5. Bhattacharjee, S., Bhattacharya, S., & Bhattacharya, A. (2020). Blood-brain barrier in brain tumours: Biology and clinical implications. Journal of Clinical Neuroscience, 78, 1–10. https://doi.org/10.1016/j.jocn.2020.04.066
6. Wen, P. Y., Weller, M., Lee, E. Q., Alexander, B. M., Barnholtz-Sloan, J. S., Barthel, F. P., … & Yung, W. K. A. (2020). Glioblastoma in adults: A Society for Neuro-Oncology (SNO) and European Society of Neuro-Oncology (EANO) consensus review on current management and future directions. Neuro-Oncology, 22(8), 1073–1113. https://doi.org/10.1093/neuonc/noaa106
7. Lenck, S., Labeyrie, M. A., Guedon, A., Saint-Maurice, J. P., Houdart, E., & Spelle, L. (2022). Advances in imaging and endovascular treatment of dural arteriovenous fistulas. Stroke, 53(2), 583–591. https://doi.org/10.1161/STROKEAHA.121.033743
8. Garcia, H. H., Nash, T. E., & Del Brutto, O. H. (2014). Clinical symptoms, diagnosis, and treatment of neurocysticercosis. The Lancet Neurology, 13(12), 1202–1215. https://doi.org/10.1016/S1474-4422(14)70094-8
9. Brook, I. (2017). Microbiology and treatment of brain abscess. Journal of Clinical Neuroscience, 38, 8–12. https://doi.org/10.1016/j.jocn.2016.12.035
10. Fischbein, N. J., & Wijman, C. A. (2010). Noncontrast cranial CT in the evaluation of acute neurologic emergencies. Neuroimaging Clinics of North America, 20(4), 579–596. https://doi.org/10.1016/j.nic.2010.07.007
11. American College of Radiology (ACR). (2024). ACR manual on contrast media (Version 11). ACR Committee on Drugs and Contrast Media. https://www.acr.org/Clinical-Resources/Contrast-Manual
12. Bae, K. T. (2010). Intravenous contrast medium administration and scan timing at CT: Considerations and approaches. Radiology, 256(1), 32–61. https://doi.org/10.1148/radiol.10090908
13. Watanabe, A., Ishido, K., Makino, K., Arita, K., Hirano, H., Tominaga, A., & Nak, M. (2021). Optimising contrast injection timing in brain CT: Impact of delay on enhancement of ring lesions in clinical neuroradiology. Neuroradiology, 63(4), 521–530. https://doi.org/10.1007/s00234-020-02597-0
14. Martinsen, A. C. T., Sæther, H. K., Hol, P. K., Olsen, D. R., & Skaane, P. (2018). Iterative reconstruction reduces abdominal CT radiation dose with preserved diagnostic quality: A systematic review and meta-analysis. European Radiology, 28(3), 1275–1283. https://doi.org/10.1007/s00330-017-5050-6
15. Bisdas, S., Shen, H., Garnon, J., Guimarães, A. R., Bongers, M. N., & Nikolaou, K. (2019). Spectral CT imaging and dual-energy CT characterisation of intracranial lesions: A comprehensive review. Insights into Imaging, 10(1), 50. https://doi.org/10.1186/s13244-019-0735-4
16. Greffier, J., Hamard, A., Pereira, F., Barrau, C., Pasquier, H., Beregi, J. P., & Frandon, J. (2020). Image quality and dose reduction opportunity of deep learning image reconstruction algorithm for CT: A phantom study. European Radiology, 30(7), 3951–3959. https://doi.org/10.1007/s00330-020-06724-w
17. Thomsen, H. S., Webb, J. A. W., & Contrast Media Safety Committee of the ESUR. (2023). Contrast media: Safety issues and ESUR guidelines (4th ed.). Springer. https://doi.org/10.1007/978-3-031-20022-5
18. van der Molen, A. J., Reimer, P., Dekkers, I. A., Bongartz, G., Bellin, M. F., Bertolotto, M., … & Stacul, F. (2018). Post-contrast acute kidney injury — Part 1: Definition, clinical features, incidence, role of contrast medium and risk factors. European Radiology, 28(7), 2845–2855. https://doi.org/10.1007/s00330-017-5246-5
19. Hittmair, K., Fleischmann, D., & Katzenbeisser, H. (2018). Temperature effects on CT contrast medium viscosity and injection dynamics: A phantom study. Journal of Computer Assisted Tomography, 42(3), 465–470. https://doi.org/10.1097/RCT.0000000000000705
20. Kawashima, H., Ogawa, F., Ohkubo, M., Matsunaga, Y., & Katada, K. (2019). Incidence of venous air embolism on contrast CT: Influence of injector-line management. Radiological Physics and Technology, 12(2), 224–230. https://doi.org/10.1007/s12194-019-00513-6
21. European Commission. (2019). Radiation protection No. 185: European guidelines on diagnostic reference levels for paediatric imaging. Publications Office of the European Union. https://doi.org/10.2833/003002
22. Söderberg, M., & Gunnarsson, M. (2010). Automatic exposure control in computed tomography — an evaluation of systems from different manufacturers. Acta Radiologica, 51(6), 625–634. https://doi.org/10.3109/02841851003698206
23. McCollough, C. H., Leng, S., Yu, L., & Fletcher, J. G. (2015). Dual- and multi-energy CT: Principles, technical approaches, and clinical applications. Radiology, 276(3), 637–653. https://doi.org/10.1148/radiol.2015142631
24. Wiggins, W. F., Goh, G. S., & Sodickson, A. D. (2022). AI-assisted CT image analysis: Current clinical applications and emerging opportunities in neuroradiology. American Journal of Neuroradiology, 43(5), 647–656. https://doi.org/10.3174/ajnr.A7500
25. Avakian, A., & Barfoot, G. (2026). Artificial intelligence in radiology: A narrative review of current methods, clinical impact, and future directions. BMC Artificial Intelligence, 2, 1. https://doi.org/10.1186/s44398-025-00020-7
26. Suh, C. H., Kim, H. S., Jung, S. C., Choi, C. G., & Kim, S. J. (2016). Clinically relevant imaging features for MR imaging of brain metastases and treatment response. American Journal of Neuroradiology, 37(8), 1421–1430. https://doi.org/10.3174/ajnr.A4749
27. Toh, C. H., Wei, K. C., Chang, C. N., Ng, S. H., Wong, H. F., & Lin, C. P. (2015). Differentiation of brain abscesses from necrotic glioblastomas and cystic metastatic brain tumors with diffusion tensor imaging. American Journal of Neuroradiology, 36(5), 906–914. https://doi.org/10.3174/ajnr.A4228
28. Mărginean, L., Ștefan, P. A., Lebovici, A., Opincariu, I., Csutak, C., Lupean, R. A., … & Suciu, B. A. (2022). CT in the differentiation of gliomas from brain metastases: The radiomics analysis of the peritumoral zone. Brain Sciences, 12(1), 109. https://doi.org/10.3390/brainsci12010109
29. Hygino da Cruz, L. C., Jr., Rodriguez, I., Domingues, R. C., Gasparetto, E. L., & Sorensen, A. G. (2011). Pseudoprogression and pseudoresponse: Imaging challenges in the assessment of posttreatment glioma. American Journal of Neuroradiology, 32(11), 1978–1985. https://doi.org/10.3174/ajnr.A2397
30. International Commission on Radiological Protection. (2020). ICRP publication 147: Use of dose quantities in radiological protection. Annals of the ICRP, 50(1). https://doi.org/10.1177/0146645320911277