Magnetic resonance imaging (MRI) can reveal structural causes of epilepsy that may be invisible from the outside: hippocampal sclerosis, focal cortical dysplasia, tumours, vascular malformations, injuries, developmental abnormalities and the consequences of previous infection or stroke. Yet access to high-quality neuroimaging is profoundly unequal across the world.
The inequality is not simply a question of whether a country owns an MRI scanner. Meaningful access depends on geography, affordability, scanner availability, maintenance, electricity, trained radiographers, neuroradiology expertise, appropriate epilepsy protocols, sedation services where needed, image quality, reporting capacity and a referral pathway capable of acting on what the scan shows.
For epilepsy, this distinction is especially important. A basic brain scan and a specialist epilepsy MRI are not always equivalent.
Neuroimaging is important — but epilepsy is not diagnosed by MRI alone
Epilepsy remains a clinical diagnosis. MRI does not show the electrical seizure itself, and a normal scan does not exclude epilepsy. Many epilepsies are genetic, network-based or otherwise not associated with a visible structural lesion on conventional imaging.
WHO guidance for non-specialist settings therefore does not require neuroimaging before treatment can begin in every person with epilepsy. Imaging is used to identify underlying pathology and refine the syndromic or aetiological diagnosis, particularly when focal-onset seizures, progressive disease, neurological abnormalities or drug resistance raise concern for a structural cause.
Where MRI and CT are both available and clinically appropriate, MRI is generally preferred for epilepsy because it provides much greater soft-tissue detail and avoids ionising radiation.
What MRI can add to epilepsy care
High-quality MRI can help identify why seizures are occurring, classify an epilepsy as structural, detect lesions associated with drug resistance, guide prognosis and treatment decisions, and determine whether further specialist assessment is needed.
In people being considered for epilepsy surgery, MRI findings are integrated with seizure semiology, EEG, neuropsychology and sometimes PET, SPECT, MEG or intracranial EEG. A clearly identified lesion that agrees with the other evidence can substantially change the route through specialist care.
The global imaging gap is not evenly distributed
MRI availability varies dramatically between countries and within countries. Reviews of global imaging capacity consistently describe large disparities between high-income and low- and middle-income settings. Some regions have dense networks of scanners, while others depend on a small number of machines serving very large populations.
Recent global health literature describes MRI scarcity in many low- and middle-income countries as a combined technological, economic and workforce problem. In practical terms, access may depend on whether a person can reach a major city, pay privately, travel across a border or wait for a scarce public appointment.
This means national scanner numbers alone can be misleading. A scanner hundreds of kilometres away, unaffordable to use, or repeatedly out of service may exist statistically while remaining inaccessible clinically.
Geography can turn availability into inaccessibility
Advanced diagnostic services are frequently concentrated in capitals and tertiary hospitals. Rural populations may face long journeys, poor transport links, accommodation costs and loss of income simply to attend a scan.
For someone with uncontrolled seizures, travel itself can also be difficult. The person may need a companion, may be unable to use certain forms of transport safely, or may need to arrange childcare and time away from work for more than one family member.
The result is a second layer of inequality: even where MRI exists, the people who live furthest from specialist services may be the least able to use it.
Cost extends far beyond the price of the scan
MRI systems are expensive to buy, install and operate. They require specialised infrastructure, regular servicing, trained personnel, quality assurance and reliable supplies. Conventional high-field scanners may also require shielding, cooling systems and carefully controlled environments.
For patients, the financial burden can include the scan fee, neurological consultation, transport, accommodation, food, time away from paid work and repeat travel if the first scan is inadequate or requires specialist review.
In health systems with high out-of-pocket spending, a technically available MRI service can therefore remain economically out of reach.
A scanner without a workforce is not a functioning service
Neuroimaging depends on people as much as machines. Radiographers must operate the scanner safely and obtain the correct sequences. Radiologists or neuroradiologists must interpret the images. Neurologists and epilepsy specialists must place the report in the clinical context. Engineers and physicists may be required to maintain performance and safety.
Shortages at any point in that chain can reduce capacity. A region may own MRI equipment yet struggle with unfilled radiology posts, limited neuroradiology expertise or insufficient training in epilepsy-specific acquisition and interpretation.
Maintenance is a global health issue
High-technology equipment can become unusable when spare parts, service contracts or specialist engineering support are unavailable. Downtime may be prolonged when components must be imported or engineers must travel internationally.
This matters because donating or purchasing a scanner is only the beginning of a sustainable imaging programme. Long-term access requires budgeting for maintenance, software, upgrades, coils, power systems, staff training and eventual replacement.
Reliable electricity and infrastructure matter
Conventional MRI systems depend on stable power and specialised physical infrastructure. In settings with unreliable electricity, voltage fluctuations or limited hospital engineering support, maintaining high-field imaging can be difficult.
Facilities also need appropriate MRI safety procedures because the magnetic field creates risks involving ferromagnetic equipment and certain implanted devices. Safe access therefore requires screening systems, staff education and controlled scanner environments.
Not every MRI is an epilepsy MRI
This is one of the most important distinctions in global epilepsy imaging. A person can have an MRI and still have a lesion missed if the acquisition was not designed for epilepsy or if the images were not reviewed with epilepsy in mind.
The International League Against Epilepsy Neuroimaging Task Force developed the HARNESS-MRI protocol to harmonise structural epilepsy imaging. Its core principle is that high-resolution, appropriately oriented three-dimensional sequences improve the search for subtle epileptogenic abnormalities.
Focal cortical dysplasia, small areas of abnormal cortical development and hippocampal abnormalities can be difficult to identify on routine brain imaging. Scanner field strength, sequence choice, slice thickness, motion and specialist interpretation all affect detection.
A normal routine MRI can therefore be misleading
‘MRI-negative epilepsy’ does not always mean there is no structural abnormality. It can mean that no lesion was visible on the study performed. Some abnormalities become apparent only after repeat imaging with a dedicated epilepsy protocol, higher-quality sequences, expert rereview or post-processing.
This creates an additional global inequality: two people may both be described as having access to MRI, while only one has access to imaging capable of answering the epilepsy-specific question.
Children face additional barriers
Paediatric neuroimaging brings extra challenges. Young children may struggle to remain still for a lengthy MRI examination, and some require sedation or anaesthesia. That introduces the need for appropriately trained teams, monitoring equipment, recovery facilities and additional safety procedures.
Brain maturation also changes the appearance of MRI across infancy and childhood, so paediatric neuroradiology expertise can be important when evaluating developmental abnormalities and early-life epilepsies.
Where sedation services or paediatric expertise are unavailable, children may experience especially long delays or receive less informative imaging.
CT remains important where MRI is unavailable
Computed tomography is faster, more widely available in many health systems and highly valuable in emergencies. It can identify haemorrhage, large tumours, hydrocephalus, calcification, major stroke, severe trauma and other urgent causes of seizures.
However, CT is less sensitive than MRI for many subtle structural causes of epilepsy and involves ionising radiation. It should not be treated as an identical substitute for specialist epilepsy MRI.
In resource-limited settings, CT may nevertheless be the only realistic imaging option. The clinical challenge is therefore not to dismiss CT, but to recognise both what it can detect and what it can miss.
Neurocysticercosis shows why imaging access can alter diagnosis
In regions where neurocysticercosis is endemic, CT and MRI may identify characteristic lesions associated with seizures and epilepsy. CT can be particularly useful for calcified lesions, while MRI can provide additional information about viable cysts, inflammation, oedema and lesion location.
Without accessible imaging, the underlying cause of seizures may remain uncertain even when the aetiology would influence clinical management and public-health understanding.
Stroke, trauma and infection create similar problems
Structural epilepsy can develop after stroke, traumatic brain injury, central nervous system infection and other acquired insults. Neuroimaging can document the injury, define its location and identify complications.
When imaging is unavailable at the time of the original illness, later epilepsy care may begin without a clear record of what happened to the brain. This can complicate classification and specialist assessment years later.
Access to imaging affects access to epilepsy surgery
Epilepsy surgery programmes depend on neuroimaging. Structural MRI is a central component of presurgical evaluation, and complex cases may require advanced imaging or image post-processing.
Where epilepsy-protocol MRI is unavailable, people with drug-resistant focal epilepsy may never reach the stage at which a surgically remediable lesion is identified. The imaging gap therefore contributes to the surgery gap.
This is especially important because drug resistance should prompt consideration of specialist evaluation rather than endless substitution of antiseizure medicines when appropriate surgical or device-based options may exist.
PET, SPECT and MEG are even more concentrated
MRI is only one layer of epilepsy imaging. Positron emission tomography (PET), ictal or interictal single-photon emission computed tomography (SPECT) and magnetoencephalography (MEG) can provide additional information in selected complex cases, especially during presurgical investigation.
These technologies require specialist equipment, radiopharmaceutical infrastructure or highly trained teams and are much less widely available than routine MRI. Their scarcity means the global diagnostic divide becomes wider as care becomes more specialised.
Image transfer can separate scanning from interpretation
Digital imaging makes it possible for scans obtained in one location to be reviewed elsewhere. Teleradiology and specialist remote review can therefore extend neuroradiology expertise beyond major centres.
This model can be particularly valuable when a country has imaging equipment but few clinicians experienced in epilepsy MRI. It also allows multidisciplinary epilepsy teams to review images alongside EEG and clinical information without requiring every expert to be physically located at the scanning hospital.
However, remote interpretation still depends on high-quality acquisition, secure data transfer, adequate bandwidth and sustainable professional capacity. A badly acquired scan cannot always be rescued by expert review.
Training can improve detection without immediately buying new scanners
Part of the access problem is technical rather than purely numerical. Training radiographers to acquire epilepsy protocols and training radiologists to recognise subtle epileptogenic lesions can improve the value of existing equipment.
Standardised protocols, structured reporting, multidisciplinary review and targeted rereading after EEG localisation are comparatively scalable ways to strengthen services while broader infrastructure develops.
Low-field and portable MRI may change the access equation
Newer low-field and portable MRI systems are being developed with lower infrastructure requirements than conventional high-field scanners. Research published in recent years has explored their potential for low-resource and remote settings, including systems that use less power, have reduced siting requirements or avoid some of the engineering demands of traditional MRI.
These technologies could expand access, but they are not automatically equivalent to high-resolution epilepsy-protocol MRI. Lower field strength can reduce signal and fine anatomical detail, and evidence for specific epilepsy applications is still developing.
The most realistic role may be complementary: widening basic neuroimaging access while preserving referral pathways to higher-resolution specialist imaging when subtle lesions are suspected.
Artificial intelligence and post-processing may help — with limits
Computer-assisted image analysis can highlight subtle abnormalities, quantify structures and support lesion detection. In epilepsy, post-processing methods are being studied and used particularly for cortical abnormalities that may be overlooked on visual inspection.
These methods could eventually help extend expertise, but they require validation across different populations, scanners and acquisition protocols. Algorithms trained on highly selected datasets from well-resourced centres may not perform identically on lower-quality or heterogeneous images from other settings.
AI should therefore be treated as decision support, not as a substitute for adequate imaging quality and clinical expertise.
Data inequality matters too
Much of the neuroimaging literature and many reference datasets have historically been generated in high-income settings. When technologies, normative databases or automated tools are developed from populations that do not represent global diversity, their generalisability may be limited.
Expanding imaging capacity in underrepresented regions is therefore important not only for clinical care but also for building more globally representative neuroscience.
Humanitarian emergencies can erase imaging access overnight
Conflict, displacement, natural disasters and health-system collapse can interrupt access to MRI even where services previously existed. Imaging facilities may be damaged, electricity disrupted, staff displaced and referral networks fragmented.
People with epilepsy may then lose both routine treatment and access to investigations needed for new neurological symptoms, injuries or worsening seizures. This is one reason continuity of epilepsy care in humanitarian settings cannot be reduced to medicine supply alone.
Waiting time is another form of restricted access
An MRI service can exist but still be functionally inaccessible if waiting lists are extremely long. Delays matter most when clinicians are trying to exclude a progressive lesion, investigate new focal neurological signs, assess drug-resistant epilepsy or plan surgery.
Capacity therefore needs to be measured not just in scanners per population but in whether clinically appropriate imaging can actually be obtained within a useful timeframe.
WHO now places diagnostic capacity within the global neurological agenda
The World Health Organization’s Intersectoral Global Action Plan on epilepsy and other neurological disorders 2022–2031 calls for stronger systems for diagnosis, treatment and care. Its implementation toolkit explicitly includes access to essential diagnostic technologies such as neuroimaging as part of service development.
This matters because the epilepsy treatment gap is not only a medication gap. It is also a diagnostic, workforce, referral and infrastructure gap.
What equitable neuroimaging access would require
Improving global access requires more than purchasing additional scanners. Sustainable systems need geographically distributed services, reliable electricity, safe facilities, maintenance contracts, spare parts, trained radiographers, radiology and neuroradiology expertise, epilepsy-specific protocols, paediatric provision, affordable financing, digital image transfer and clear referral pathways.
Health systems also need to know when a routine scan is sufficient and when a person should be referred for dedicated epilepsy MRI or advanced presurgical imaging.
The objective is not that every hospital should possess every form of neuroimaging. It is that a person with epilepsy should be able to reach the level of imaging their clinical situation requires.
What this page does — and does not — explain
This page is about worldwide access to MRI and neuroimaging. The technical role of MRI in individual epilepsy assessment is covered on MRI and Neuroimaging in Epilepsy. The difference between routine imaging and specialist protocols is covered on Why a Dedicated Epilepsy MRI Is Different. PET, SPECT and MEG are covered separately in the presurgical investigation section.
Sources
World Health Organization — Role of neuroimaging in management of epilepsy
WHO — Intersectoral Global Action Plan implementation toolkit
International League Against Epilepsy Neuroimaging Task Force — Structural MRI recommendations and HARNESS-MRI
Murali et al. — Bringing MRI to low- and middle-income countries: directions, challenges and potential solutions
MRI scarcity in low- and middle-income countries
Neuroimaging in Low- to Middle-Income Countries: A Health Equity Perspective
Low field, high impact: democratizing MRI for clinical and research innovation
Evidence checked: September 2026.