How Epilepsy Is Diagnosed

By Dovy Paukstys, Founder, Komori Care

This article is educational and is not medical advice. It does not diagnose, treat, prevent, or cure any condition. Komori is not a medical device, is not FDA-cleared, does not detect seizures, and is not intended for people with epilepsy. Talk with a neurologist or epileptologist about personal medical decisions. Households living with epilepsy should use clinically validated seizure-detection devices as directed by their care team.

Nobody diagnoses epilepsy from a comment thread. I will die on that hill. ;)

I don't have epilepsy. I also don't read EEGs. I opened the NINDS epilepsy page and the CDC treatment page so I could write the actual sequence clinicians use, not the TV version.

Diagnosis is a clinician's call. Usually a neurologist. Sometimes an epileptologist. History first. Then tests. Then a label, or not.

The everyday rule NINDS still leads with

NINDS: generally, epilepsy is diagnosed after two or more unprovoked seizures separated by at least 24 hours.

That's the sentence most people hear first. Accurate diagnosis matters because treatment hangs on it. Having a seizure does not, by itself, mean epilepsy. First seizures, febrile seizures, nonepileptic events, eclampsia. Those can involve seizures or seizure-like episodes and still not be epilepsy.

CDC's basics page: epilepsy is a brain disorder that causes repeated seizures. A seizure is a change in normal brain activity that lasts from a few seconds to a few minutes. They also say less than half of newly diagnosed cases have a known cause.

Two unprovoked events is the public version. A clinician can still decide something else in a specific case. I'm not doing that from here.

History is still the main tool

NINDS calls a detailed medical history, including symptoms and duration, one of the best methods they have for deciding what kind of seizure happened and what kind of epilepsy it might be. Past illnesses. Other symptoms. Family history of seizures.

People who have a seizure often don't remember it. So witness accounts matter. A lot. The person gets asked whether anything weird happened first (a warning, an aura). Observers get asked for a timeline. What the body did. How long. What the person was like after.

Phone video from a family member has saved more clinic visits than people think. Still not a diagnosis. It's a better history.

If you were the only one in the room, say that. Unknown onset is a real category. Pretending you saw the start when you didn't just wastes everyone's time.

EEG

Electroencephalogram. Measures electrical activity in the brain. NINDS: it can look for abnormal brain waves and may help decide if antiseizure drugs would help.

A normal EEG does not automatically mean "not epilepsy." An abnormal EEG does not automatically mean "epilepsy." Clinicians know that. Internet commenters pretend they don't.

Video-EEG is the same test with cameras. Used to see what the events actually are, and to rule out disorders that look like epilepsy. That's the tool for "are these even epileptic."

Some people later get longer monitoring in a hospital unit. Some get stereo-EEG, which is electrodes inside the brain, when a surgery question is on the table. MEG reads magnetic signals from neurons. Those are specialist tools. Not step one for a first stare spell.

MRI and the other pictures

CT and MRI look for structural problems. Tumors. Cysts. Things that can cause seizures. Functional MRI maps normal activity and can show function problems. PET looks at chemical activity and can show low-metabolism regions after a seizure. SPECT can compare blood flow during a seizure with blood flow between seizures, in a hospital monitoring setting.

CDC's treatment page says it more simply. Doctors look for causes so they can identify the type of epilepsy and the treatment. They prescribe medicine or something else if needed. They refer to a specialist if needed.

MRI is not a personality test. A clean MRI is common. NINDS already said about half of people never get a named cause.

Blood tests and the rest

Blood work can screen for metabolic or genetic problems that contribute. It can also look for infection, lead poisoning, anemia, diabetes. Things that provoke or mimic.

Developmental, neurological, and behavioral tests measure motor skill, behavior, intellect. They show how epilepsy is affecting the person and can hint at the type. They don't replace the seizure history.

Nonepileptic events exist. NINDS: they look like seizures without the electrical discharge. Psychogenic nonepileptic seizures don't respond to antiseizure drugs. Other lookalikes include narcolepsy, Tourette, cardiac arrhythmia. That's why video-EEG exists. Guessing from the hallway is how people get the wrong drug.

Who should be in the room

CDC: ER doctors, family doctors, and pediatricians often see the first seizure. They may diagnose or they may refer. Specialists are neurologists (brain and nervous system) and epileptologists (neurologists who specialize in epilepsy).

See a specialist if seizures aren't controlled on medicine, or the side effects are wrecking you. An epilepsy center is for hard-to-control seizures or specialized care. Their teams can include epileptologists, neurosurgeons, specialized nurses, psychologists, social workers, EEG techs.

If events are new, changing, or unexplained, that's a clinician, same day or next clinic, not a blog FAQ.

What this page is not

Not a self-score. Not "if you have 3 of these 8 symptoms." Not a consumer-device readout. A wearable or a camera does not diagnose epilepsy. Komori doesn't either. Not a medical device. Doesn't detect seizures. Not intended for people with epilepsy.

Households that already have an epilepsy diagnosis and need detection tools should use what the care team validated and directed. That's a different sentence from "an app said it was a seizure."

FAQ

Can one seizure mean epilepsy?

Not automatically. NINDS: having a seizure does not by itself mean a person has epilepsy. The everyday diagnostic line they publish is two or more unprovoked seizures at least 24 hours apart. A clinician decides.

What tests are used?

History and witness accounts first. Then often EEG, sometimes video-EEG, MRI or CT, blood tests, and in selected cases PET, SPECT, MEG, or invasive monitoring. NINDS lists those. Not everyone gets all of them.

Does a normal EEG rule epilepsy out?

No. NINDS describes EEG as a tool that can find abnormal waves and help with treatment questions. Clinicians still use the history. A normal tracing is not a trophy.

Who diagnoses it?

A physician. Often a neurologist or epileptologist. CDC says first-contact doctors may diagnose or refer. No website, app, or consumer monitor gets a vote.

Sources

• NINDS, Epilepsy and Seizures: https://www.ninds.nih.gov/health-information/disorders/epilepsy-and-seizures
• CDC, Epilepsy Basics: https://www.cdc.gov/epilepsy/about/index.html
• CDC, Treatment of Epilepsy: https://www.cdc.gov/epilepsy/treatment/index.html