A Wireless EEG Patch for Continuous Electrographic Monitoring

Objectives: The primary objectives of this study are to evaluate the devices ability: (1) to record wireless EEG from patients in the long-term EEG monitoring unit (2) compare wireless EEG in these recordings to standard-of-care wired EEG (3) to record wireless EEG from patients undergoing ambulatory EEG monitoring (4) compare wireless EEG in these recordings to the ambulatory EEG findings (5) develop automated seizure detection algorithms based on the data collected herein (6) develop real-time seizure alerting based on the data collected herein (7) record longitudinal EEG from Epilog outside of the epilepsy monitoring unit while patients answer daily questions through a mobile app designed to provide a probability of the patient having a seizure event. Hypotheses: Single-channel EEG recorded with Epilog can be used to accurately detect seizures. Single-channel EEG recorded with Epilog can be used to alert to seizures in real time. Single-channel EEG recorded with Epilog can be used along with psychological, behavioral, and environmental input to a mobile app to create an hourly forecast of seizure probability. Background: Continuous electroencephalography (EEG) is a relatively expensive, time-consuming, inpatient monitoring procedure that can require several days to complete. It is designed to aid epileptologists in the diagnosis and localization of seizures and other abnormalities. EEG monitoring outside of the clinic is limited to bulky and cumbersome ambulatory monitoring systems that are prone to a variety of noise artifacts outside of the controlled confines of the long-term EEG monitoring (LTM) unit. An effective wireless EEG solution could improve long-term (continuous) EEG recordings. Importantly, long-term EEG monitoring outside of the hospital could yield important data regarding the general brain state, including seizure occurrence, in a realistic at-home setting. There is a clear need for a small, easy to use, reliable, wireless device for human EEG monitoring to aid in the treatment of a patients seizure disorder. Patients are frequently referred to the long-term epilepsy monitoring unit (EMU) in order to evaluate a suspect underlying neurological condition that may manifest through EEG and is a common procedure. Despite its clinical significance and the desire to minimize the time spent in the hospital, there is a gap in the ability to maintain a quantifiable report of EEG activity in the patients home environment. In general, patients diagnosed with a seizure disorder are asked to keep a seizure diary that is designed to aid the clinician in diagnosis. Patient self reporting of seizures with a seizure diary is often unreliable up to 55% of all seizures can be missed, and still others go misreported [1-3]. This is especially true of the developmentally delayed population where the burden of self-reporting seizures is often placed upon a care giver. Most seizures are non-convulsive and can be relatively subtle. It is extremely important for effective treatment to know how an anticonvulsant is affecting a patients seizures, particularly focal seizures. Often, seizures are amnesic and unless someone else recognizes the clinical signs of the seizure, it may go unnoticed[4]. Many seizures can occur at night or while a patient is sleepy or stressed [5-7] further making the detection of seizures difficult. Obtaining an accurate report of such unrecognizable seizures is difficult for persons with epilepsy and their families. Early detection of acute electrographic seizures could lead to more-effective application of therapeutic approaches to block or minimize brain damage. Over 30% of patients with epilepsy have intractable seizures that are difficult to manage, are an immense financial burden to the person with epilepsy and their family, and drastically reduce quality of life [8, 9]. Thus, there remains a need to obtain quantifiable, continuous EEG records after a diagnosis of epilepsy. This study will allow the development of automated seizure detection, alerting, and prediction algorithms based on Epitel's Epilog single-channel, wireless EEG sensor. This proposal will evaluate an individual-use, patch-type telemetry sensor for simplified, single-channel EEG recording from human patients. This sensor can be applied in a simple fashion to the scalp allowing real-time recording of the EEG signal on the sensor. This relatively small unit, roughly the size of a thick Band-Aid is self-contained, water resistant, and powered by a watch battery. The sensor is compact and would be applied and/or shown how to use during a clinical visit. The sensor is noninvasive and is attached to the patient by external placement on intact skin below hairline with a one-piece, one-day-use disposable conductive-adhesive sticker. Research Design and Methods: Study design: This is both a prospective, un-blinded, observational study as well as a prospective un-blinded interventional study. For the inpatient unit, the findings of standard wired EEG will be compared to the findings of wireless EEG recording device to develop a training dataset for machine learning algorithms for automated seizure detection. For the outpatient unit, the characteristics of the EEG recorded with be combined with psychological, behavioral, and environmental input by the patient in a mobile app that then provides an hourly probability (percent change) of having a seizure in the next hour. Study population: The proposed patient population for this study involves pediatric patients, ages 5-17. The consideration of the pediatric population as a vulnerable population for participation in this study has been taken into account in determining the appropriateness of this protocol. This study will specifically provide information as to the specific needs of the pediatric patient population; which is anticipated to be a significant percentage of the user population of the proposed device. The risk assessment of the study anticipates that the pediatric subject population will not be exposed to risks greater than the current standard of care for which they are already being evaluated. Subject Selection: Inclusion Criteria for inpatient and outpatient unit: Previous diagnosis of Epilepsy or other seizure disorder admitted for scheduled EEG testing, 5 to 17 years of age, inclusive Exclusion criteria for the inpatient and outpatient unit: Known or suspected allergy to adhesives Study methods: Up to four Epilog single-channel EEG sensors will be placed alongside gold-standard wired EEG in the epilepsy monitoring unit. Each sensor (1" x 1" x 1/4", 6.6 g) is attached to the scalp below hairline with a one-piece conductive adhesive sticker consisting of a hypoalergenic hydrocolloid and hydrogel. Sensors are placed in roughly the F7, F8, TP9, or TP10 electrode locations in the international 10-10 system. Epilog is constructed of a circuit board with built-in gold EEG recording electrodes attached to both ends. The sensors are then be sealed and packaged in the form of a pliable medical-grade silicone, and be designed to have a low profile, such that it will not be easily dislodged or be uncomfortable to the user. The wireless EEG device is powered by a single lithium watch battery (CR2016, 270mW) sealed within the sensor to IPX3. The Epilog sensors only use batteries approved to IEC-60086-4. Epilog transmits EEG through secure Bluetooth when interacting with the mobile app. Otherwise, Epilog is capable of storing 10 consecutive days of EEG when the Bluetooth radio is disabled. Once activated, the sensor begins transmitting and/or data logging the signal. The clinician can connect the device to the USB port of an HIPAA-approved, password protected laptop or other computer device for activating the sensor and/or imputing patient study ID information in the sensor and/or downloaded data off of the sensor according to the European Data Format, an internationally recognized standard EEG format (www.edfplus.info). In order to visualize the EEG signal in real time, a specifically-designed, mobile app can be used when Bluetooth is enabled on each sensor. For outpatient use, the ability to view EEG is disabled. Data can be downloaded from the device by connecting to an USB-enabled, password-protected computer. An USB-based acquisition hardware, computer, and software with an EDF-file viewer will be used for viewing, analyzing, and storing of EEG data according to HIPAA patient privacy policies. Device information: Labeling: All investigational devices will have the following label statement: CAUTION Investigational Device. Limited by Federal law to investigational use. Each sensor will also be labeled with the manufacturer, business location, quantity of contents and for identification and traceability, by serial number and date of manufacture. Classification of the sensor: This sensor is suitably similar to FDA-cleared, electroencephalography devices under section number 21 CFR 882.1400, identified as a device used to measure and record electrical activity of the patients brain obtained by placing two or more electrodes on the head. This sensor is subject to FDA Class II performance standards. This sensor route-to-market is through 510(k) clearance by demonstration of substantial equivalence to a legally marketed predicate device. Sensor Receipt: Epitel will directly deliver sensors to the Principal Investigator, who will log them on the sensor accountability form in the Study Binder. Sensor Storage: Sensors are singly housed in boxes marked with the sensor serial number and a label statement: CAUTION Investigational Device. Limited by Federal law to investigational use. Sensors will be kept under material control in the epilepsy monitoring unit prior to and after use. Sensors shall be stored at approximately room temperature, under ambient conditions. There are no other special storage requirements. Sensor Dispensing: Sensors will be logged on a device accountability form in the Study Binder when they 1 are delivered to neurology, 2 are placed on a study patient, 3 data has been downloaded from the device, 4 device data has been erased, and 5 devices have been returned to and received by the Study Sponsor. This will be signed and dated by the study team. Sensor Disposition: After use, the study team will return sensors to the storage box/cabinet after downloading the data. Data on the sensors will be erased and each sensor will be used once if in-patient and only on a single patient if out-patient. Sensors will then be returned to the Study Sponsor where they will be accounted for, destroyed, and properly disposed of. Return or Destruction of Sensors: Following collection of the data and sensor inspection, used sensors stored at neurology with battery level below 1.4V will be returned to the Study Sponsor where they will be accounted for, destroyed, and properly disposed of. Recruitment and consent process in inpatient and outpatient units: General Consent Processes: Patients presenting to the EMU for standard-of-care procedures will be recruited for this study. The Study Coordinator also maintains the responsibility of contacting eligible patients physician to ask for permission to approach family and patient via email. If the physician grants permission, the Study Coordinator will approach and inform the patient/guardian of the opportunity to participate in this study. All participants will be required to participate in the consent/assent process. The patient and legal guardians will be walked through the informed consent and assent documents and will be given appropriate time to ask questions on the study. Consent/assent process for all patients who meet the criteria for enrollment: The study will be described verbally when the Study Coordinator approaches the patient and family and the description will also be written in the consent document. The patients will be given a chance to ask questions at any point during the enrollment process. Upon their agreement to participate in the study, they will be asked to sign the consent form and give verbal consent/assent. There are no external recruitment procedures for this study. Study Procedures: Screening process: Patients will be screened by the research staff prior to their hospital admission. The research staff team will review the weekly admission list and screen patients for enrollment appropriateness. Once an eligible patient is selected, the research team will call or email the patients asummary of the study and request approval to approach the patient for consent/assent when they are admitted. Enrollment: Once eligibility is determined and the research team receives approval to approach, the eligible patient and their family will be given a verbal description of the study by the research staff and offered a chance to participate in the study. The eligible patient and their family where appropriate, will be walked through the informed consent documents and will be given appropriate time to ask questions on the study. After informed consent to participate in the study is obtained, the patient will be enrolled. Patient enrollment may be performed by any of the study team members, as listed in the roles/responsibilities section. Patient enrollment is documented in the Study Binder. Sensor Placement: Patients will be monitored by the Epilog sensor(s) and full-montage wired EEG for patients who remain in the EMU. Sensors will be placed per the neurologists direction, next to a standard electrode in an area where seizures are known to occur in the event that epilepsy has been previously diagnosed or suspect to occur. Written documentation on the placement location(s) of each sensor alongside the standard wired EEG electrodes shall be recorded for each patient. Patients will wear Epilog for as long as wired EEG is simultaneously recorded (1-7 days in the EMU). For patients who are to wear Epilog continuously outside of the EMU, instructions will be provided in both written, a verbal interactive demonstration, as well as video of how to use Epilog and the mobile app. After use, the Epilog sensors will be removed by gently peeling it off the scalp. For out-patient use, patients and their family/caregivers are instructed to replace the conductive-adhesive sticker once per day, typically while showering or bathing, or before bed. Risks and Benefits: Study Risks: The risks of this study are minimal. There is a potential risk that the patient may have an adverse reaction to the conductive-adhesive gel used to fix the recording electrodes to the scalp. This risk will be minimized by observing the scalp around the device daily for redness or irritation that could indicate a reaction to the conductive-adhesive. Only FDA-clear hypoalergenic conductive hydrogel and adhesive hydrocolloids are used. Study Benefits: There are no direct benefits to the patient for taking part in this study. The potential benefit to clinics and patients longer term is that recording wireless EEG may prove useful for determining if a single channel of wireless, long-term EEG can be used to detect, alert, and predict seizure activity outside of the hospital during a patients everyday activities. The EEG may also be used to determine the prevalence of normal versus abnormal sleep cycling in patients, though this information will not be directly evaluated in this study. Withdrawal: Subject withdrawal: Subjects will exit the study if any of the following conditions exist; 1) Subject voluntarily withdraws from the study, 2) Subjects well-being, in the opinion of the Investigator would be compromised by study continuation, 3) Subject experiences an adverse event defined by the Investigator as device related. How to withdraw subjects: A subject may report their voluntary withdrawal to any of the contacts of the study. Any study team member receiving a notification of withdrawal shall report this to the Study Coordinator. If an Investigator withdraws a subject; this shall be documented in the Study Binder with the reason for withdrawal. If there were any adverse events related to the withdrawal, the Investigator and the Study Coordinator in their role as Safety Monitor are responsible for investigating, documenting and if applicable, reporting these incidents. The Study Coordinator and Principal Investigator of the study are responsible for ensuring the safety of the patient in the event of withdrawal and that all dispensed devices have been recovered. The withdrawal of any patient is documented in the Study Binder under participant information. Safety: There are no risks to the patient for premature withdrawal from this study. Follow-up: A reasonable effort shall be made to follow-up with all study participants who withdraw from the study after the time of withdrawal to ensure that there are no device related adverse events. Minimally, this effort should obtain survival data on the patient. The Study Coordinator is responsible for this follow-up. Study Data: Upon enrollment the following data will be collected as part of the study: sex, age, documentation of no known skin allergy/sensitivity to adhesives (exclusionary characteristic), history of epilepsy diagnosis (if applicable). During the Study, wireless EEG will be obtained from the study device. For in-patient, the EEG will be recorded by traditional wired procedure for the duration the study device is worn. For out-patient data, daily responses to ta set of survey questions presented to the patient by the Aura app, modified with PROMIS, developed by Dr. Madison Berl. Each person enrolled will answer baseline survey questions that include (1) average number of hours one sleeps per-day during a typical week, and (2) 10-question Perceived Stress Scale survey. Answers to these baseline questions will personalize risk associated with each change from baseline over time. As the out-patient study begins and enrollees wear Epilog for the 90 day period, the mobile app will continue to ask behavioral and psychological survey questions as a means to effect a change in the seizure forecast provided by the mobile app. The number of questions asked of the user by the app will not exceed four per day to reduce burden of using the app on a daily basis. Along with physiological parameters (EEG and circadian seizure susceptibility), an hourly seizure forecast will be estimated and optimized post hoc. This would be a forecast the user would see and could use to plan their day. Importantly, we will track how the user interacts with the mobile app to determine adherence, how often the app is used, clicks, crash statistics, and overall satisfaction. It may be the case where the patient enrolled begins to answer questions posed by the app infrequently or not at all. To mitigate this problem the app will continue to prompt the user to answer the questions if they are dismissed. If the user misses two or more days, the study coordinator will contact the patient to encourage adherence. Analysis plan: The primary outcome measures for this study are seizure event times marked by epileptologists as the start, end, artifacts, and seizure type during the in-patient studies. These data are used to develop the automated seizure detection and alerting algorithms. For out-patient studies, the patients' interaction with the mobile app, how often the app is used, clicks, crash statistics, and overall satisfaction with the seizure forecast probability performance. References: Blum, D.E., et al., Patient awareness of seizures. Neurology, 1996. 47(1): p. 260-4. Hoppe, C., A. Poepel, and C.E. Elger, Epilepsy: accuracy of patient seizure counts. Arch Neurol, 2007. 64(11): p. 1595-9. 3. Poochikian-Sarkissian, S., et al., Patient awareness of seizures as documented in the epilepsy monitoring unit. Can J Neurosci Nurs, 2009. 31(4): p. 22-3. 4. Fisher, R.S., et al., Seizure diaries for clinical research and practice: limitations and future prospects. Epilepsy Behav, 2012. 24(3): p. 304-10. 5. Derry, C.P. and S. Duncan, Sleep and epilepsy. Epilepsy Behav, 2013. 26(3): p. 394-404. 6. Matos, G., et al., Sleep and epilepsy: exploring an intriguing relationship with a translational approach. Epilepsy Behav, 2013. 26(3): p. 405-9. 7. Novakova, B., et al., The role of stress as a trigger for epileptic seizures: a narrative review of evidence from human and animal studies. Epilepsia, 2013. 54(11): p. 1866-76. 8. Luoni, C., et al., Determinants of health-related quality of life in pharmacoresistant epilepsy: results from a large multicenter study of consecutively enrolled patients using validated quantitative assessments. Epilepsia, 2011. 52(12): p. 2181-91. 9. Schuele, S.U. and H.O. Luders, Intractable epilepsy: management and therapeutic alternatives. Lancet Neurol, 2008. 7(6): p. 514-24. 10. Freeman, W.J., et al., Spatial spectra of scalp EEG and EMG from awake humans. Clin Neurophysiol, 2003. 114(6): p. 1053-68. 11. Davey, M.P., J.D. Victor, and N.D. Schiff, Power spectra and coherence in the EEG of a vegetative patient with severe asymmetric brain damage. Clin Neurophysiol, 2000. 111(11): p. 1949-54