An electroencephalography is only recommended in those who likely had an epileptic seizure and may help determine the type of seizure or syndrome present. In children it is typically only needed after a second seizure. It cannot be used to rule out the diagnosis and may be falsely positive in those without the disease. In certain situations it may be useful to prefer the EEG while sleeping or sleep deprived.

Diagnostic imaging by CT scan and MRI is recommended after a first non-febrile seizure to detect structural problems inside the brain. MRI is generally a better imaging test except when intracranial bleeding is suspected. Imaging may be done at a later point in time in those who return to their normal selves while in the emergency room. If a person has a previous diagnosis of epilepsy with previous imaging repeat imaging is not usually needed with subsequent seizures.

In adults, testing electrolytes, blood glucose and calcium levels is important to rule these out as causes, as is an electrocardiogram. A lumbar puncture may be useful to diagnose a central nervous system infection but is not routinely needed. Routine antiseizure medical levels in the blood are not required in adults or children. In children additional tests may be required.

A high blood prolactin level within the first 20 minutes following a seizure may be useful to confirm an epileptic seizure as opposed to psychogenic non-epileptic seizure. Serum prolactin level is less useful for detecting partial seizures. If it is normal an epileptic seizure is still possible and a serum prolactin does not separate epileptic seizures from syncope. It is not recommended as a routine part of diagnosis epilepsy.

Differentiating an epileptic seizure from other conditions such as syncope can be difficult. Other possible conditions that can mimic a seizure include: decerebrate posturing, psychogenic seizures, tetanus, dystonia, migraine headaches, and strychnine poisoning. In addition, 5% of people with a positive tilt table test may have seizure-like activity that seems to be due to cerebral hypoxia. Convulsions may occur due to psychological reasons and this is known as a psychogenic non-epileptic seizure. Non-epileptic seizures may also occur due to a number of other reasons.

An electroencephalogram (EEG) can assist in showing brain activity suggestive of an increased risk of seizures. It is only recommended for those who are likely to have had an epileptic seizure on the basis of symptoms. In the diagnosis of epilepsy, electroencephalography may help distinguish the type of seizure or syndrome present. In children it is typically only needed after a second seizure. It cannot be used to rule out the diagnosis and may be falsely positive in those without the disease. In certain situations it may be useful to perform the EEG while the affected individual is sleeping or sleep deprived.

Diagnostic imaging by CT scan and MRI is recommended after a first non-febrile seizure to detect structural problems in and around the brain. MRI is generally a better imaging test except when bleeding is suspected, for which CT is more sensitive and more easily available. If someone attends the emergency room with a seizure but returns to normal quickly, imaging tests may be done at a later point. If a person has a previous diagnosis of epilepsy with previous imaging, repeating the imaging is usually not needed even if there are subsequent seizures.

For adults, the testing of electrolyte, blood glucose and calcium levels is important to rule out problems with these as causes. An electrocardiogram can rule out problems with the rhythm of the heart. A lumbar puncture may be useful to diagnose a central nervous system infection but is not routinely needed. In children additional tests may be required such as urine biochemistry and blood testing looking for metabolic disorders.

A high blood prolactin level within the first 20 minutes following a seizure may be useful to help confirm an epileptic seizure as opposed to psychogenic non-epileptic seizure. Serum prolactin level is less useful for detecting focal seizures. If it is normal an epileptic seizure is still possible and a serum prolactin does not separate epileptic seizures from syncope. It is not recommended as a routine part of the diagnosis of epilepsy.

The diagnosis can be confirmed when the characteristic centrotemporal spikes are seen on electroencephalography (EEG). Typically, high-voltage spikes followed by slow waves are seen. Given the nocturnal activity, a sleep EEG can often be helpful. Technically, the label "benign" can only be confirmed if the child's development continues to be normal during follow-up. Neuroimaging, usually with an MRI scan, is only advised for cases with atypical presentation or atypical findings on clinical examination or EEG.

The disorder should be differentiated from several other conditions, especially centrotemporal spikes without seizures, centrotemporal spikes with local brain pathology, central spikes in Rett syndrome and fragile X syndrome, malignant Rolandic epilepsy, temporal lobe epilepsy and Landau-Kleffner syndrome.

Diagnosis of epilepsy can be difficult. A number of other conditions may present very similar signs and symptoms to seizures, including syncope, hyperventilation, migraines, narcolepsy, panic attacks and psychogenic non-epileptic seizures (PNES). In particular a syncope can be accompanied by a short episode of convulsions. Nocturnal frontal lobe epilepsy, often misdiagnosed as nightmares, was considered to be a parasomnia but later identified to be an epilepsy syndrome. Attacks of the movement disorder paroxysmal dyskinesia may be taken for epileptic seizures. The cause of a drop attack can be, among many others, an atonic seizure.

Children may have behaviors that are easily mistaken for epileptic seizures but are not. These include breath-holding spells, bed wetting, night terrors, tics and shudder attacks. Gastroesophageal reflux may cause arching of the back and twisting of the head to the side in infants, which may be mistaken for tonic-clonic seizures.

Misdiagnosis is frequent (occurring in about 5 to 30% of cases). Different studies showed that in many cases seizure-like attacks in apparent treatment-resistant epilepsy have a cardiovascular cause. Approximately 20% of the people seen at epilepsy clinics have PNES and of those who have PNES about 10% also have epilepsy; separating the two based on the seizure episode alone without further testing is often difficult.

Diagnosis may be made by noting the correlation between exposure to specific visual stimuli and seizure activity. More precise investigation can be carried out by combining an EEG with a device producing "Intermittent Photic Stimulation" (IPS). The IPS device produces specific types of stimuli that can be controlled and adjusted with precision. The testing physician adjusts the IPS device and looks for characteristic anomalies in the EEG, such as photoparoxysmal response (PPR), that are consistent with PSE and/or may herald the onset of seizure activity. The testing is halted before a seizure actually occurs.

Sometimes diagnostic indicators consistent with PSE can be found through provocative testing with IPS, and yet no seizures may ever occur in real-life situations. Many people will show PSE-like abnormalities in brain activity with sufficiently aggressive stimulation, but they never experience seizures and are not considered to have PSE.

The prognosis of ICOE-G is unclear, although available data indicate that remission occurs in 50–60% of patients within 2–4 years of onset. Seizures show a dramatically good response to carbamazepine in more than 90% of patients. However, 40–50% of patients may continue to have visual seizures and infrequent secondarily generalized convulsions, particularly if they have not been appropriately treated with antiepileptic drugs.

The differential diagnosis of ICOE-G is mainly from symptomatic occipital epilepsy and migraine where misdiagnosis is high. The differential diagnosis from migraine should be easy because elementary visual hallucinations of occipital seizures develop rapidly within seconds, are brief in duration (2–3 minutes) are usually colored and circular. These are fundamentally different from the visual aura of migraine which develops slowly in minutes, is longer lasting ≥5 minutes and mainly achromatic with linear patterns.

Symptomatic occipital epilepsy often imitates ICOE-G; neuroophthalmological examination and brain imaging may be normal. Thus, high resolution MRI is required to detect subtle lesions.

The differentiation of ICOE-G from Panayiotopoulos syndrome is straightforward. The seizures of ICOE-G are purely occipital, brief, frequent and diurnal. Conversely seizures in Panayiotopoulos syndrome manifest with autonomic manifestations, they are lengthy and infrequent; visual symptoms are rare and not the sole manifestation of a seizure.

Diagnosis is made upon history of absence seizures during early childhood and the observation of ~3 Hz spike-and-wave discharges on an EEG.

The most important factor in diagnosing a patient with vertiginous epilepsy is the subject’s detailed description of the episode. However, due to the associated symptoms of the syndrome a subject may have difficulty remembering the specifics of the experience. This makes it difficult for a physician to confirm the diagnosis with absolute certainty. A questionnaire may be used to help patients, especially children, describe their symptoms. Clinicians may also consult family members for assistance in diagnosis, relying on their observations to help understand the episodes. In addition to the description of the event, neurological, physical and hematologic examinations are completed to assist in diagnosis. For proper diagnosis, an otological exam (examination of the ear) should also be completed to rule out disorders of the inner ear, which could also be responsible for manifestations of vertigo. This may include an audiological assessment and vestibular function test. During diagnosis, history-taking is essential in determining possible causes of vertiginous epilepsy as well as tracking the progress of the disorder over time.

Other means used in diagnosis of vertiginous epilepsy include:

- Electroencephalography (EEG)

- Magnetic resonance imaging (MRI)

- Positron emission tomography (PET)

- Neuropsychological testing

The EEG measures electrical activity in the brain, allowing a physician to identify any unusual patterns. While EEGs are good for identifying abnormal brain activity is it not helpful in localizing where the seizure originates because they spread so quickly across the brain. MRIs are used to look for masses or lesions in the temporal lobe of the brain, indicating possible tumors or cancer as the cause of the seizures. When using a PET scan, a physician is looking to detect abnormal blood flow and glucose metabolism in the brain, which is visible between seizures, to indicate the region of origin.

The diagnosis or suspicion of LGS is often a question of probability rather than certainty. This is because the varied presentations of LGS share features with other disorders, many of which may be said to have overlapping characteristics.

The diagnosis is more obvious when the epilepsy has frequent and manifold attacks, with the classic pattern on the electro-encephalogram (EEG); the latter is a slowed rhythm with Spike-wave-pattern, or with a multifocal and generalizing Sharp-slow-wave-discharges at 1.5–2.5 Hz. During sleep, frequently, tonic patterns can be seen. But variations of these patterns are known in patients with no diagnosis other than LGS, and they can differ bilaterally, and from time to time, within the same patient.

General medical investigation usually reveals developmental delay and cognitive deficiencies in children with true LGS. These may precede development of seizures, or require up to two years after the seizures begin, in order to become apparent.

Exclusion of organic or structural brain lesions is also important in establishing a correct diagnosis of LGS; this may require magnetic resonance imaging (MRI) or computerized tomography (CT). An important differential diagnosis is 'Pseudo-Lennox-Syndrome', which differs from LGS, in that there are no tonic seizures; sleeping EEG provides the best basis for distinguishing between the two.

Intravenous immunoglobulin therapy has been used in Lennox–Gastaut syndrome as early as 1986, when van Rijckevorsel-Harmant and colleagues used it in seven patients with ostensibly idiopathic LGS and saw EEG improvement and decreased seizure frequency in six of them.

As with video games, rapidly changing images or highly regular patterns such as flashing banner ads or irregular fonts can trigger seizures in people with photosensitive epilepsy. Two sets of guidelines exist to help web designers produce content that is safe for people with photosensitive epilepsy:

- The World Wide Web Consortium - Web Content Accessibility Guidelines (WCAG) Version 2.0, produced in 2008, specifies that content should not flash more than 3 times in any 1 second period. However it does allow flashing above this rate if the flashing is below the "general and red flashing thresholds". (Basically, it is OK to flash more than 3 times in a 1-second period if the flashing is small enough or low contrast enough.)

- In the United States, websites provided by federal agencies are governed by section 508 of the Rehabilitation Act. The Act says that pages shall be designed to avoid causing the screen to flicker with a frequency greater than 2 Hz and less than 55 Hz. The 508 regulations are currently being updated and are expected to use the same criteria as WCAG 2.0 when finished.

- A free tool for evaluating Web Content for flashing called the Photosensitive Epilepsy Analysis Tool (PEAT) is available from the Trace R&D Center at the University of Maryland.

The differential diagnosis of PNES firstly involves ruling out epilepsy as the cause of the seizure episodes, along with other organic causes of non-epileptic seizures, including syncope, migraine, vertigo, anoxia, hypoglycemia, and stroke. However, between 5-20% of patients with PNES also have epilepsy. Frontal lobe seizures can be mistaken for PNES, though these tend to have shorter duration, stereotyped patterns of movements and occurrence during sleep. Next, an exclusion of factitious disorder (a subconscious somatic symptom disorder, where seizures are caused by psychological reasons) and malingering (simulating seizures intentionally for conscious personal gain – such as monetary compensation or avoidance of criminal punishment) is conducted. Finally other psychiatric conditions that may superficially resemble seizures are eliminated, including panic disorder, schizophrenia, and depersonalisation disorder.

The most conclusive test to distinguish epilepsy from PNES is long term video-EEG monitoring, with the aim of capturing one or two episodes on both videotape and EEG simultaneously (some clinicians may use suggestion to attempt to trigger an episode). Conventional EEG may not be particularly helpful because of a high false-positive rate for abnormal findings in the general population, but also of abnormal findings in patients with some of the psychiatric disorders that can mimic PNES. Additional diagnostic criteria are usually considered when diagnosing PNES from long term video-EEG monitoring because frontal lobe epilepsy may be undetectable with surface EEGs.

Following most tonic-clonic or complex partial epileptic seizures, blood levels of serum prolactin rise, which can be detected by laboratory testing if a sample is taken in the right time window. However, due to false positives and variability in results this test is relied upon less frequently.

Some features are more or less likely to suggest PNES but they are not conclusive and should be considered within the broader clinical picture. Features that are common in PNES but rarer in epilepsy include: biting the tip of the tongue, seizures lasting more than 2 minutes (easiest factor to distinguish), seizures having a gradual onset, a fluctuating course of disease severity, the eyes being closed during a seizure, and side to side head movements. Features that are uncommon in PNES include automatisms (automatic complex movements during the seizure), severe tongue biting, biting the inside of the mouth, and incontinence.

If a patient with suspected PNES has an episode during a clinical examination, there are a number of signs that can be elicited to help support or refute the diagnosis of PNES. Compared to patients with epilepsy, patients with PNES will tend to resist having their eyes forced open (if they are closed during the seizure), will stop their hands from hitting their own face if the hand is dropped over the head, and will fixate their eyes in a way suggesting an absence of neurological interference. Mellers et al. warn that such tests are neither conclusive nor impossible for a determined patient with factitious disorder to "pass" through faking convincingly.

The prognosis for Rolandic seizures is invariably excellent, with probably less than 2% risk of developing absence seizures and less often GTCS in adult life.

Remission usually occurs within 2–4 years from onset and before the age of 16 years. The total number of seizures is low, the majority of patients having fewer than 10 seizures; 10–20% have just a single seizure. About 10–20% may have frequent seizures, but these also remit with age.

Children with Rolandic seizures may develop usually mild and reversible linguistic, cognitive and behavioural abnormalities during the active phase of the disease. These may be worse in children with onset of seizures before 8 years of age, high rate of occurrence and multifocal EEG spikes.

The development, social adaptation and occupations of adults with a previous history of Rolandic seizures were found normal.

Criteria for diagnosis of abdominal epilepsy includes frequent periodic abdominal symptoms, an abnormal electroencephalogram (EEG) and significant improvement of gastrointestinal symptoms after taking anti-seizure medication. Medical testing for diagnosis can be completed using MRI scans of the brain, CT scans and ultrasounds of the abdomen, endoscopy of the gastrointestinal tract, and blood tests.

The trigger needs to be identified before prescribing anti-epileptics. The most commonly prescribed drugs for reflex epilepsy are valproate, carbamazepine and clonazepam, though lamotrigine, levetiracetam are promising.

Diagnosis is typically made upon patient history, although EEG recordings can be confirmatory if they occur during attacks.

In epilepsy surgery a distinction can be made between resective and disconnective procedures. In a resective procedure the area of the brain that causes the seizures is removed. In a disconnective procedure the neural connections in the brain that allow the seizures to spread are disconnected. In most cases epilepsy surgery is only an option when the area of the brain that causes the seizures - the so-called epileptic focus can be clearly identified and is not responsible for critical functions such as language. Several imaging techniques such as magnetic resonance tomography and functional techniques like electrocorticography are used to demarcate the epileptic focus clearly.

Epilepsy surgery has been performed since the 1860s and doctors have observed that it is highly effective in producing freedom from seizures. However, it was not until 2001 that a scientifically sound study was carried out to examine the effectiveness of temporal lobectomy.

Temporal lobe surgery can be complicated by decreased cognitive function. However, after temporal lobectomy, memory function is supported by the opposite temporal lobe; and recruitment of the frontal lobe. Cognitive rehabilitation may also help.

Childhood absence epilepsy is a fairly common disorder with a prevalence of 1 in 1000 people. Few of these people will likely have mutations in CACNA1H or GABRG2 as the prevalence of those in the studies presented is 10% or less.

The test is particularly indicated in children who have had cluster seizures in series. It is also recommended for patients who are diagnosed GEFS+ and when the seizures are associated with fever, infection, experienced regression, delayed cognitive growth or behavioral problems. The test is typically ordered by neurologists. The diagnostic test can be done by drawing blood or saliva of the patient and their immediate family. It is analyzed in laboratories that specialize in genetic testing. Genetic testing can aid in a firmer diagnosis and understanding of the disorder, may aid in identifying the optimal treatment plan and if positive, testing of the parents can determine if they are carriers. (See Genetic Counseling)

PCDH19 gene-related epilepsy is clinically based on patient and family seizure history, cognitive and behavioral neuropsychological evaluation, neurological examination, electroencephalogram (EEG) studies, and long term observation. Diagnosis is confirmed using molecular testing for PCDH19 mutations.

Most children who develop epilepsy are treated conventionally with anticonvulsants. In about 70% of cases of childhood epilepsy, medication can completely control seizures. Unfortunately, medications come with an extensive list of side effects that range from mild discomfort to major cognitive impairment. Usually, the adverse cognitive effects are ablated following dose reduction or cessation of the drug.

Medicating a child is not always easy. Many pills are made only to be swallowed, which can be difficult for a child. For some medications, chewable versions do exist.

The ketogenic diet is used to treat children who have not responded successfully to other treatments. This diet is low in carbohydrates, adequate in protein and high in fat. It has proven successful in two thirds of epilepsy cases.

In some cases, severe epilepsy is treated with the hemispherectomy, a drastic surgical procedure in which part or all of one of the hemispheres of the brain is removed.