What Is Clinical Neurophysiology and What Is It Used For?

Clinical neurophysiology is the medical specialty that uses electrical recordings of the nervous system to diagnose conditions, measure their severity, and track how well treatments are working.¹ If you have ever had electrodes placed on your scalp to check for seizures, had a nerve conduction study to investigate tingling in your hand, or been monitored during spinal surgery, you have encountered clinical neurophysiology in action. The field spans a surprisingly broad range of conditions, from epilepsy and multiple sclerosis to sleep apnea and treatment-resistant depression, and the tools keep expanding.

How Electrical Recordings Reveal What Imaging Cannot

An MRI or CT scan shows you what the brain and nerves look like. Clinical neurophysiology shows you how they are functioning in real time. A structurally normal-looking nerve can still conduct signals too slowly. A brain that looks perfectly fine on a scan can be generating abnormal electrical discharges that cause seizures. The recordings generated in a clinical neurophysiology lab capture timing, speed, and patterns of electrical activity, and those details often point to a diagnosis that structural imaging would miss entirely.

The tests range from completely non-invasive (electrodes stuck to the skin with paste) to mildly uncomfortable (a thin needle inserted into a muscle) to procedures performed under general anesthesia (electrodes implanted directly in the brain for epilepsy mapping). What ties them all together is the same basic principle: nerves and muscles communicate using electrical signals, and by recording those signals at the right place and time, a trained specialist can spot where the system is breaking down.

Electroencephalography and Seizure Diagnosis

The most widely known tool in clinical neurophysiology is the electroencephalogram, or EEG. Electrodes placed along the scalp pick up the brain’s electrical activity, and the resulting tracings are read by a neurophysiologist looking for abnormal patterns. EEG remains the standard method for identifying seizures because it is low-cost and non-invasive, and the gold standard for seizure identification is still visual recognition of abnormal electrical patterns by a trained specialist.²

A routine EEG lasts about 20 to 40 minutes, but that snapshot is not always enough. In intensive care settings, continuous EEG monitoring can run for hours or days to catch seizures that have no visible physical signs. These non-convulsive seizures are easy to miss because the patient may simply appear confused or unresponsive rather than showing the dramatic convulsions most people associate with epilepsy. In one study of neurosurgical patients monitored with continuous EEG, seizure activity was sometimes detected only after the first 30 minutes of recording, and subtle electrographic seizure patterns were initially overlooked even by reviewers.³ Newer headset-style continuous video EEG systems designed for the ICU have shown high specificity for detecting non-convulsive status epilepticus, helping clinicians act faster when a patient’s brain is seizing silently.⁴

Nerve Conduction Studies and Electromyography

If EEG is the brain-focused side of the field, nerve conduction studies (NCS) and electromyography (EMG) are its peripheral counterpart. NCS measures how fast and how strongly electrical signals travel along a nerve, while EMG records the electrical activity of muscles at rest and during contraction. Together they can pinpoint whether a problem lies in the nerve, the muscle, or the junction between the two.

These tests are the go-to investigation for conditions like carpal tunnel syndrome, radiculopathy caused by a compressed spinal nerve root, and peripheral neuropathy from diabetes or other causes. The procedure is generally well tolerated. NCS involves small electrical pulses delivered through surface electrodes, which feel like a brief static shock. EMG involves a thin needle electrode inserted into the muscle, which can be uncomfortable but is rarely described as truly painful. In the hands of a trained practitioner, serious side effects are rare, though the needle portion does carry a small theoretical risk of bleeding, infection, or nerve injury.^{5 6}

One increasingly common companion to NCS and EMG is neuromuscular ultrasound. Traditional electrodiagnostic tests tell you how a nerve is functioning, but they do not show you what the nerve looks like physically. Ultrasound fills that gap, providing real-time images of nerves, muscles, and surrounding structures. It has been studied extensively in the assessment of entrapment neuropathies and complements electrodiagnostic studies by supplying anatomic detail that electrical recordings alone cannot provide.^{7 8}

Evoked Potentials and Multiple Sclerosis

Evoked potentials are recordings of the electrical signals the brain generates in response to a specific stimulus, such as a flashing checkerboard pattern (visual evoked potentials), a clicking sound (auditory evoked potentials), or a mild electrical pulse applied to a limb (somatosensory evoked potentials). The key measurement is latency, or how long it takes the signal to travel from the stimulus to the brain. A delayed response suggests the nerve pathway is damaged or that the insulating myelin sheath surrounding it has broken down.

This makes evoked potentials particularly valuable in multiple sclerosis, where the immune system strips away myelin. Visual evoked potentials (VEPs) are routinely used to evaluate demyelination or axonal damage.⁹ A study comparing VEP results in MS patients with healthy subjects found a significant difference in the latency of the P100 wave, and tracking VEPs over time provided information about disease progression and treatment response.¹⁰ Somatosensory evoked potentials add another layer. In a study of 105 MS patients, over 80% of those with definite or early probable MS had abnormal somatosensory evoked potential results, even when clinical signs were subtle.¹¹

The practical upshot for patients is that evoked potentials can detect damage the person does not yet feel. If a visual pathway is conducting slowly but the patient has no noticeable vision problems, that subclinical finding can help confirm a diagnosis or prompt earlier treatment.

Protecting the Nervous System During Surgery

Intraoperative neurophysiological monitoring, or IONM, is one of the field’s most consequential applications. During surgeries that put the spinal cord, nerve roots, or brain at risk, a neurophysiology team continuously records electrical signals to catch damage as it is happening, rather than discovering it after the patient wakes up. The surgeon receives real-time alerts, giving them the chance to adjust the procedure before an injury becomes permanent.

IONM is most commonly used in spinal surgery, where it has been adopted widely to reduce the incidence of postoperative neurological complications involving the spinal cord, cauda equina, and nerve roots.¹² Evidence supports the use of multimodal monitoring, which combines several recording techniques simultaneously, to improve neurological outcomes.¹³ A nine-year review of children who underwent non-spine surgery but had spinal cords at risk found that IONM allowed the surgical team to identify and react to alerts that may have otherwise led to permanent neurological injury.¹⁴

IONM is not limited to spinal procedures. It is also used during brain tumor surgery, thyroid surgery (to protect the recurrent laryngeal nerve and avoid voice damage), and surgeries near major peripheral nerves. The monitoring team is typically led by a clinical neurophysiologist or a trained technologist who interprets the signals in real time.

Diagnosing Neuromuscular Junction Disorders

Myasthenia gravis is a condition where the immune system attacks the connection between nerves and muscles, causing fluctuating weakness. Diagnosing it can be tricky because symptoms wax and wane, and blood tests are not always positive. Clinical neurophysiology offers a specialized test called single-fiber electromyography (SFEMG), which is considered the most sensitive electrodiagnostic tool for detecting this type of transmission problem.

In a prospective study of 100 patients, SFEMG had a sensitivity of about 98% for diagnosing myasthenia gravis, meaning it correctly identified the disease in nearly all confirmed cases.¹⁵ Its specificity in that study was around 70%, which means some patients without the disease may test positive. A related technique using a concentric needle rather than a single-fiber electrode showed somewhat lower sensitivity but higher specificity, making it useful for confirming the diagnosis when SFEMG raises the suspicion.¹⁶ These two techniques together give clinicians a way to both screen for and confirm neuromuscular junction problems.

Sleep Studies

Polysomnography, the formal name for a clinical sleep study, is firmly rooted in clinical neurophysiology. It is the gold standard diagnostic method for sleep disorders, recording electrophysiological signals including brain waves, eye movements, chin muscle tone, heart rhythm, breathing effort, airflow, and blood oxygen levels throughout the night.¹⁷

A sleep study is most commonly ordered to evaluate obstructive sleep apnea, but it also helps diagnose narcolepsy, restless legs syndrome, periodic limb movement disorder, parasomnias like sleepwalking, and REM sleep behavior disorder (a condition sometimes linked to later development of Parkinson’s disease). The report generated by polysomnography, detailing sleep stages, breathing events, and limb movements, guides treatment decisions ranging from CPAP therapy to medication changes.¹⁸

Electrophysiology in Movement Disorders

Tremor, myoclonus, and other involuntary movements are often diagnosed clinically, by observation alone. But the human eye has limits. Electrophysiological tests can measure things clinical examination cannot reliably assess on its own, including tremor frequency, the rhythmicity of a tremor, and whether tremors in different body parts are being driven by the same central generator or by separate ones.¹⁹

One of the most practically important applications is distinguishing organic tremors from functional (psychogenic) tremors. In functional tremor, the involuntary movement is real and distressing, but it is not caused by a structural neurological disease. Electrophysiology can help tell the two apart by checking whether a tremor can be entrained or suppressed by voluntary movements of the opposite limb, or by analyzing the frequency and coherence patterns of muscle signals.²⁰ The distinction matters enormously for treatment. A patient with functional tremor benefits from physiotherapy and psychological approaches, while a patient with essential tremor or Parkinson’s tremor needs a different treatment pathway entirely. Similarly, reflex circuits like the blink reflex can help differentiate organic blepharospasm from its functional counterpart, because the recovery cycle of the blink reflex is abnormally enhanced in the organic form but normal in the functional one.²¹

Autonomic Nervous System Testing

Clinical neurophysiology does not stop at the brain, spinal cord, and limb nerves. It also covers the autonomic nervous system, which controls involuntary functions like heart rate, blood pressure regulation, sweating, and digestion. When this system fails, people experience dizziness on standing, fainting, abnormal sweating, or bowel and bladder dysfunction. The challenge is that these symptoms overlap with many other conditions, so objective testing is needed.

An international expert consensus recommends that proper autonomic testing combines assessments of at least three domains: sympathetic cardiovascular (adrenergic) function, parasympathetic cardiovagal function, and sudomotor (sweat gland) function, because no single test alone is sufficient to diagnose the degree or distribution of autonomic failure.²² Common tests include heart rate variability during deep breathing, blood pressure response to tilting the patient upright, and quantitative sudomotor axon reflex testing (a sweat test). The results can help identify and grade conditions like diabetic autonomic neuropathy, pure autonomic failure, and multiple system atrophy.

Transcranial Magnetic Stimulation as a Diagnostic and Therapeutic Tool

Transcranial magnetic stimulation (TMS) straddles the boundary between clinical neurophysiology and treatment. In its diagnostic role, TMS uses a magnetic coil held against the scalp to generate a brief pulse that activates brain cells. By recording the resulting muscle twitch, a neurophysiologist can measure how quickly signals travel from the brain to the muscles, helping identify problems with the corticospinal tract, the main highway connecting the brain to the body’s motor system.

In its therapeutic role, repetitive TMS (rTMS) delivers trains of pulses to specific brain regions to alter their activity over time. The evidence is strongest for treatment-resistant depression, where rTMS has become an established option when medications have not worked.²³ Research is also exploring TMS for post-stroke recovery and Alzheimer’s disease, where it has demonstrated effects on brain plasticity.²⁴ A systematic review concluded that TMS has viable neuromodulatory potential and can be employed as a therapy alternative for a range of neuropsychiatric disorders.²⁵

Neonatal and Pediatric Applications

Babies who experience oxygen deprivation during birth need rapid assessment of brain function, and standard EEG setups are impractical in a neonatal intensive care unit. A simplified version called the amplitude-integrated EEG (aEEG) has filled this gap. It compresses and filters the EEG signal into a trend display that can be monitored at the bedside for hours or days. Its main uses in newborns are early evaluation of brain function after birth asphyxia and detection of seizures.²⁶

A systematic review of aEEG’s seizure detection ability found that when the compressed trend was used together with a raw EEG trace, sensitivity for identifying individual seizures reached a median of about 76%, with specificity around 85%. Without the raw trace, sensitivity dropped considerably.²⁷ The practical takeaway is that aEEG is a valuable screening tool in the NICU, but it works best when clinicians also have access to the underlying raw signal for confirmation.

When Surface Recordings Are Not Enough

For the roughly one-third of epilepsy patients whose seizures are not controlled by medication, surgery to remove the seizure focus may be an option. But identifying that focus with certainty sometimes requires going beyond surface EEG. Stereoelectroencephalography, or SEEG, uses thin depth electrodes implanted directly into the brain through small holes in the skull. These electrodes can sample both superficial and deep structures, and even both hemispheres simultaneously, creating a three-dimensional map of where seizures originate and how they spread.²⁸

SEEG has become the preferred method for intracranial EEG monitoring at most comprehensive epilepsy centers in North America, largely because it carries lower complication rates than the older approach of placing large electrode grids on the brain’s surface, and because it can reach areas those grids cannot, like the insula and the cingulate gyrus deep in the brain’s folds.²⁹ The recording data guide the surgical team in deciding exactly what brain tissue to remove or disconnect.

Magnetoencephalography

Magnetoencephalography (MEG) records the tiny magnetic fields generated by electrical currents in the brain, rather than the electrical voltages that EEG picks up. The practical advantage is that magnetic signals pass through the skull without distortion, which gives MEG better spatial precision for locating the source of abnormal brain activity. In epilepsy, MEG is probably more sensitive than EEG at detecting certain types of interictal spikes, particularly in areas like the superficial frontal cortex and the lateral temporal neocortex, though the two techniques are usually complementary rather than one replacing the other.³⁰

The main limitation is cost and availability. MEG requires a magnetically shielded room and extremely sensitive detectors, making it available only at specialized centers. It is typically reserved for presurgical epilepsy evaluation when the question of where exactly seizures start has not been answered by other tests, or for localizing brain regions responsible for language and motor function before surgery.

Artificial Intelligence and the Future of Neurophysiology

Reading EEG, EMG, and other neurophysiological recordings is labor-intensive and requires specialized training. A single 24-hour continuous EEG recording can produce thousands of pages of data, and trained specialists are in short supply. Artificial intelligence is beginning to change that landscape. AI-based EEG analysis uses machine learning and deep learning to automate seizure detection, real-time analysis, and data quality improvement, significantly reducing the workload of human interpreters and improving diagnostic efficiency.³¹

One study comparing machine learning approaches for epileptic seizure detection found that a deep learning model achieved a validation accuracy of 97%.³² These numbers are promising, but the field is not yet at a point where algorithms replace human judgment. The current role of AI is as a screening layer: flagging suspicious segments so a neurophysiologist can focus their attention where it matters most, rather than scrolling through hours of normal background activity. As wearable EEG devices become smaller and more practical, AI-assisted interpretation will likely become essential for handling the volume of data they generate outside the hospital setting.