Frequently Asked Questions

Get the answers to some of the most commonly asked questions about Ceribell’s AI-powered point-of-care EEG

Point-of-care EEG (POC EEG) refers to portable EEG systems that non-specialist healthcare providers can use to initiate brain monitoring at the bedside. The Ceribell Point-of-Care EEG system is designed to make brain monitoring fast, easy, and accessible. Any trained provider can bring the portable EEG machine, a disposable headband and handheld recorder, into the patient’s room, apply it in under five minutes1, and start continuously recording brain activity. The system uses the Clarity™ AI algorithm to monitor for seizure activity in real time and provides bedside alerts within minutes, helping clinicians make faster, more confident treatment decisions when time is brain.

While traditional EEG machines are designed for outpatient evaluation, using them in the hospital often requires waiting hours for a trained EEG technician to arrive and set up the system1,2. Additional delays can occur while waiting hours to days for the EEG to be formally interpreted3,4. Unlike a conventional EEG, Ceribell does not require an EEG technician for setup because any trained provider can apply the system and start monitoring brain activity in minutes. Ceribell’s Clarity AI algorithm continuously monitors for non-convulsive seizure activity and automatically alerts providers when prolonged seizure activity is detected, helping them quickly recognize when a patient may need urgent intervention.

Ceribell’s portable POC EEG system is used in over 500 hospitals across the U.S., from large academic medical centers to small rural community hospitals. To date, more than 200,000 patients have been assessed using Ceribell. Users are frontline providers who need rapid brain monitoring to inform critical care decisions including ICU teams, emergency physicians, neurologists, hospitalists, stroke teams, intensivists, and nurses.

Many of the approximately three million annual acute care patients in the U.S. are at risk of neurological conditions,5,6 including non-convulsive seizures that can only be detected with an EEG. In fact, up to 92% of seizures in the critical care setting are non-convulsive.7,8 If not detected and treated quickly, this type of seizure can lead to ongoing brain injury, long-term disability, or even death9,10, making fast, accurate brain monitoring essential for timely diagnosis and intervention.

The Ceribell POC EEG monitoring device addresses the critical need for on-demand brain monitoring, by enabling any trained provider to rapidly apply an EEG and initiate monitoring for suspected seizures. It combines a handheld recorder and headband with a powerful AI algorithm that continuously monitors brain activity in real time. Designed for ease of use at the bedside, Ceribell supports fast diagnosis, targeted treatment, improved patient outcomes, and reduced length of stay1,2,11,12, all without straining clinical resources.

Clarity is Ceribell’s proprietary AI algorithm that interprets EEG signals captured by the portable EEG system. Built on over 800,000 hours of expert-annotated EEG recordings, Clarity continuously monitors for seizures, analyzing data every 10 seconds. If sustained seizure activity is detected, Clarity automatically generates a visual and audible alert at the bedside.

Clarity was trained using thousands of EEG recordings from Ceribell’s proprietary database, ensuring maximum accuracy. While competitors often rely on third-party algorithms, Clarity was built and validated on Ceribell’s portable EEG hardware.

The algorithm assesses a comprehensive set of EEG features to continuously monitor brain activity. Based on these assessments, Clarity generates a rolling seizure burden measurement, which is critical for informing real-time clinical decision-making.

Seizure burden refers to the cumulative amount of time a patient is experiencing seizure activity relative to the total duration of EEG monitoring. Clinically, a high seizure burden is associated with an increased risk of brain injury and long-term disability.10,13,14

The Ceribell system measures seizure burden in real time, displaying the percentage of time the patient is in seizure over a rolling five-minute window on the bedside recorder. This allows clinicians to assess brain activity quickly and easily.

When seizure burden reaches 90% or higher over a five-minute period, the Ceribell system generates an automatic visual and audible alert, helping the bedside team act promptly to review the EEG and initiate appropriate treatment.

Clarity is the only AI algorithm designed specifically for Ceribell’s portable EEG hardware and validated on large, real-world datasets. Many competing systems generally rely on third-party algorithms with limited validation and several missed cases of status epilepticus. In contrast, Clarity’s 7th-generation algorithm demonstrates consistently high sensitivity (95%) and specificity (97%) to status epilepticus.15

In 2023, Clarity became the first and only AI software cleared by the FDA to diagnose electrographic status epilepticus (ESE) at the bedside.16

No. Ceribell headbands are single patient use and disposable, which eliminates the need for disinfection between patients. This design aligns with hospital infection prevention protocols and helps reduce the risk of cross-contamination, unlike reusable EEG headsets or helmets that must be carefully disinfected after every use. The Ceribell recorder can be disinfected if necessary, using standard sani-wipes.

Ceribell is the evidence leader for point-of-care EEG when it comes to clinical evidence. The system has been validated in more than 35 peer-reviewed publications and in multiple multicenter clinical studies including DECIDE1, SAFER12,17, and AccuRASE18, which assessed more than 1,000 patients. Overall, there are more than 110 publications and abstracts demonstrating Ceribell’s technical performance, clinical impact, workflow efficiency, and economic value. This evidence shows that Ceribell improves diagnostic accuracy, reduces ICU and hospital length of stay, and helps optimize patient outcomes.

Yes. Ceribell qualifies for existing EEG CPT codes and follows the same reimbursement approach as conventional EEG. In addition to standard reimbursement for EEG, CMS granted Ceribell a New Technology Add-on Payment code for Clarity which provides for additional potential reimbursement of up to $913.90 per eligible patient.19

No. Ceribell is a diagnostic tool, not a treatment device. It helps healthcare providers detect and monitor seizure activity in real time, empowering faster and more informed treatment decisions. Providers use the information generated by Ceribell to guide appropriate anti-seizure medication administration.

Non-convulsive seizures (NCS) are often present with minimal to no outward symptoms other than an unexplained change in mental status, lethargy, or coma.20 They are also referred to as sub-clinical seizures. NCS are highly prevalent in neurocritical care patients,7-10 yet constitute a poorly understood neurological emergency. Like convulsive seizures, without prompt management with anti-seizure medication, non-convulsive seizures can become more difficult to manage, progress to status epilepticus and lead to increased rates of cognitive morbidity and mortality.21,22 The only way that NCS can be diagnosed is using EEG to monitor the disruption in electrical patterns in the brain.

Status epilepticus (SE) is a prolonged period of continuous clinical and/or electrographic seizure activity, or recurrent seizure activity without recovery (returning to baseline) between seizures, which can have long-term consequences (after certain time point).23,24 When there is no physical/clinical correlate to the prolonged seizures, they may also be referred to as non-convulsive status epilepticus (NCSE). Electrographic status epilepticus (ESE) is a term used by the American Clinical Neurophysiology Society to define prolonged electrographic seizures visible on EEG for ≥ 10 minutes or 20% of an hour).25 Regardless of the specific name, if SE is left uncontrolled, negative outcomes (including potential permanent disability or death) are significantly increased.26,27

An increasing number of societies recognize non-convulsive seizure as a significant risk that occurs frequently in patients with serious underlying conditions and requires EEG for prompt diagnosis and management. Read the recommendations from The Neurocritical Care Society, The Joint Commission, the American Heart Association and others here.

Yes. The Ceribell system with the Clarity seizure detection algorithm is FDA-cleared for use in patients aged 1 year and older*. For patients under 1 year old, the system can be used without the Clarity algorithm. The headband itself is cleared for use in patients of any age, provided it fits.

*Clarity™ for pediatric use is FDA-cleared, but not yet commercially available.

References and Citations

1. Vespa, P., et al. (2020) Crit Care Med. 48(9):1249-1257 
2. Gururangan, K., et al. (2016) Clinical Neurophysiology. 127(10):3335-3340 
3. Quigg, M., et al. (2001) J Clin Neurophysiol. 18(2):162-165 
4. Gavvala, J., et al. (2014) Epilepsia. 55(11):1864-1871  
5. Liberman, A.L., et al. (2023) JAMA Neurol. 80(2):213-215  
6. Moodley, K.K., et al. (2018) Clin Med (Lond). 18(5):418-421 
7. Rudin, D., et al. (2011) Epilepsy Res. 96(1-2):140-150 
8. Claassen, J., et al. (2004). Neurology. 62(10):1743-1748 
9. Laccheo, I., et al. (2015) Neurocrit Care. 22(2):202-211  
10. De Marchis, G.M., et al. (2016) Neurology. 86(3):253-260 
11. Yazbeck, M., et al. (2019) J Neurosci Nurs. 51(6):308-312 
12. Desai, M., et al. (2024) Neurocrit Care.  
13. Payne, E.T., et al. (2014) Brain 137(5):1429-1438 
14. Gutierrez-Viedma, A., et al. (2021) Acta Neruol Scand. 143:281-289 

15. Karunakaran, S., et al. (2024) AES2024 Annual Meeting Poster​ 
16. FDA 510k Clearance Letter K223504  
17. Kalkach-Aparicio, M., et al. (2024) Neurology. 103(2): e209621  
18. Sheikh, Z., et al. (2024) Neurology 104(2) 
19. CMS FY 2024 IPPS Final Rule. CMS-1785-F, 88 Fed. Reg. 58927-58930, Aug. 28, 2023  
20. Fatma, T., et al. (2016) J. King Saud Univ. – Comput. Inf. Sci. 28(4):407-415 
21. Young, G.B., et al. (1996) Neurology. 47(1):83-89 
22. Lowenstein, D. H., et al. (1993) Neurology. 43(3 Pt 1):483-488 
23. Brophy, G., et al. (2012) Neurocrit Care. 17(1):3-23 
24. Trinka, E., et al. (2015) Epilepsia. 56(10):1515-1523 
25. Hirsch, L.J., et al. (2021) J Clin Neurophysiol. 38(1):1-29 
26. Trinka, E., et al. (2015) Epilepsia. 56(10):1515-1523 
27. Hillman, J., et al. (2013) Int J Emerg Med.6(1):6