(192 days)
Unknown
The document describes "automated algorithms" for signal analysis and event recognition, but does not explicitly mention AI, ML, or specific techniques like deep learning. While these automated algorithms could be based on AI/ML, the summary does not provide enough detail to confirm this.
No.
The device is intended for diagnostic evaluation to assess sleep quality and aid in the diagnosis of sleep and respiratory-related sleep disorders, not for therapeutic purposes.
Yes
Explanation: The "Intended Use / Indications for Use" section explicitly states that "EnsoSleep is intended for use for the diagnostic evaluation by a physician to assess sleep quality and as an aid for the diagnosis of sleep and respiratory related sleep disorders in adults only." This clearly indicates its diagnostic purpose.
Yes
The device description explicitly states "EnsoSleep is a software-only medical device". It analyzes previously recorded physiological signals and operates as a software application.
Based on the provided information, EnsoSleep is not an In Vitro Diagnostic (IVD) device. Here's why:
- IVD Definition: In Vitro Diagnostics are devices intended for use in the collection, preparation, and examination of specimens taken from the human body (such as blood, urine, or tissue) to provide information for diagnostic purposes.
- EnsoSleep's Function: EnsoSleep analyzes previously recorded physiological signals obtained during sleep. These signals are not specimens taken from the human body for in vitro examination. Instead, they are recordings of physiological activity (like brain waves, muscle activity, and airflow) captured by external sensors during a sleep study.
- Input Data: The input data for EnsoSleep is EDF or EDF+ files, which are standard formats for storing physiological data from polysomnography (PSG) studies. This is distinct from the biological specimens used in IVD testing.
- Intended Use: While EnsoSleep is used as an aid for the diagnosis of sleep disorders, it does so by analyzing physiological signals, not by examining biological specimens. The diagnosis is ultimately made by a physician based on the analysis provided by EnsoSleep and other clinical information.
Therefore, EnsoSleep falls under the category of a medical device that analyzes physiological data, rather than an In Vitro Diagnostic device.
N/A
Intended Use / Indications for Use
EnsoSleep is intended for use for the diagnostic evaluation by a physician to assess sleep quality and as an aid for the diagnosis of sleep and respiratory related sleep disorders in adults only. EnsoSleep is a software-only medical device to be used under the supervision of a clinician to analyze physiological signals and automatically score sleep study results, including the staging of sleep, detection of arousals, leg movements, and sleep disordered breathing events including obstructive apneas. All automatically scored events are subject to verification by a qualified clinician. Central apneas, mixed apneas, and hypopneas must be manually marked within records.
Product codes (comma separated list FDA assigned to the subject device)
OLZ
Device Description
EnsoSleep is a software application that analyzes previously recorded physiological signals obtained during sleep. The EnsoSleep software can analyze any EDF or EDF+ files.
Automated algorithms are applied to the raw signals in order to derive additional signals and interpret the raw and derived signal information. The software automates recognition of:
Sleep Stage Events
- Wake
- Stage N1
- Stage N2
- · Stage N3
- Stage REM
Respiratory Events
- · Sleep disordered breathing (apneas and hypopneas)
- · Apneas detected with airflow signal are classified as obstructive apnea (OSA), and can be edited to be central or mixed appeas
- · Sleep disordered breathing events not detected to be apneas are marked as hypopnea
- · Central apneas, mixed apneas, and hypopneas must be manually marked within records
Arousal Events
· Arousals
- Movement Events
- · Periodic Leg Movements during Sleep (PLMS)
The EnsoSleep software can be used as a stand-alone application for use on Microsoft Windows 7 & 8 operating system platforms. All processing, scoring, and analysis of signal data occurs on the EnsoSleep cloud servers.
Mentions image processing
Not Found
Mentions AI, DNN, or ML
Not Found
Input Imaging Modality
Not Found
Anatomical Site
Not Found
Indicated Patient Age Range
Adults only
Intended User / Care Setting
Physician, Clinician / Sleep study laboratory
Description of the training set, sample size, data source, and annotation protocol
Not Found
Description of the test set, sample size, data source, and annotation protocol
Clinical performance testing was completed by evaluating EnsoSleep device performance using a cross-sectional experimental design on a representative N=72 subject sample of retrospective clinical PSG data. First, the intended use population, study population, conditions of interest, designated comparative reference, designated comparative benchmarks, and experimental endpoints were defined. Upon applying predefined selection controls, a statistically representative sample of the defined intended use and user population sample consisting of N=22 archived collection of retrospective diagnostic clinical PSG data collected from an AASM Accredited. The study population was then sent to a clinical testing laboratory where each PSG was manually scored by three (3) independent registered sleep technologists (RPSGT) that met all acquisition, scoring-blind, and rater controls. A designated comparative reference was constructed using 2/3 Majority Scoring to evaluate the EnsoSleep device performance versus the predicate Sleep Profiler and MICHELE device performance benchmarks and acceptance criteria of positive percent agreement (PA), negative percent agreement (NA), and overall percent (OA) were predefined competitively based on analysis of sleep staging event detection and diagnostic agreement performance reported in the predicate device 510(k) documentation. The predicate device did not report clinical testing results for staging, arousal, and leg movement event detection agreement performance, and therefore the reference device clinical performance and 510(k) documentation were used to facilitate a valid comparison.
Summary of Performance Studies (study type, sample size, AUC, MRMC, standalone performance, key results)
Substantial equivalence was established through a testing protocol that used clinical polysomnography (PSG) data to evaluate the performance of EnsoSleep using a cross-sectional experimental design on a representative N=72 subject sample of retrospective clinical PSG data.
Endpoint 1: Sleep Staging Performance (N=72 subjects, 59719 epochs for EnsoSleep vs 2/3 Majority Scoring; N=43 subjects, 31361 epochs for Sleep Profiler (K153412) vs 2/3 Majority Scoring). EnsoSleep PA, NA, and OA performance was observed to show no statistically significant differences or was observed to be significantly greater in all comparisons relative to the predicate device.
Key Results for EnsoSleep (Median Percent Agreement with 95% CI):
- Wake: PA 86% (82%, 88%), NA 97% (95%, 98%), OA 94% (92%, 95%)
- N1: PA 41% (33%, 48%), NA 94% (93%, 96%), OA 91% (90%, 93%)
- N2: PA 77% (73%, 81%), NA 87% (85%, 90%), OA 83% (80%, 85%)
- N3: PA 81% (74%, 88%), NA 93% (91%, 95%), OA 92% (90%, 94%)
- REM: PA 79% (72%, 84%), NA 99% (98%, 99%), OA 96% (96%, 97%)
- Total: PA 78% (77%, 80%), NA 95% (94%, 95%), OA 91% (91%, 92%)
Endpoint 2: Sleep Apnea Diagnostic Agreement (N=72 for EnsoSleep; N=60 for Sleep Profiler (K153412) for overall AHI; N=40 for Sleep Profiler (K153412) for REM AHI). EnsoSleep PA and NA performance for all diagnostic agreements (mild-REM, moderate-overall, and moderate-REM) showed no statistically significant differences compared to predicate device performance.
Key Results for EnsoSleep (Median Percent Agreement with 95% CI) and Likelihood Ratios:
- AHI >= 5: PA 91% (82%, 98%), NA 76% (61%, 90%), OA 85% (77%, 92%), LR(+) 3.76, LR(-) 0.12
- AHI >= 15: PA 95% (83%, 100%), NA 98% (94%, 100%), OA 97% (93%, 100%), LR(+) 52.25, LR(-) 0.05
- REM AHI >= 5: PA 83% (72%, 94%), NA 89% (79%, 97%), OA 86% (79%, 93%), LR(+) 7.71, LR(-) 0.19
- REM AHI >= 15: PA 79% (56%, 94%), NA 96% (90%, 100%), OA 92% (85%, 97%), LR(+) 22.00, LR(-) 0.22
Endpoint 3: Event Detection Agreement (N=72 subjects, 59719 epochs for EnsoSleep vs 2/3 Majority Scoring; N=30 subjects, 24987 epochs for MICHELE (K112102) vs 2/3 Majority Scoring). Point-estimates of EnsoSleep PA and NA event detection performance exceeded, were equivalent to, or were within 10% of the reference performance.
Key Results for EnsoSleep (Median Percent Agreement with 95% CI):
- Sleep Disordered Breathing Events: PA 67% (58%, 75%), NA 93% (92%, 94%), OA 91% (90%, 92%)
- Apnea Events: PA 56% (41%, 70%), NA 96% (96%, 97%), OA 95% (95%, 96%)
- Obstructive Apnea Events: PA 53% (35%, 71%), NA 97% (96%, 97%), OA 96% (95%, 97%)
- Arousal Events: PA 66% (61%, 71%), NA 90% (88%, 91%), OA 87% (85%, 88%)
- Leg Movement Events: PA 71% (60%, 80%), NA 90% (89%, 92%), OA 89% (87%, 90%)
Overall, performance test results demonstrated that EnsoSleep staging event detection, sleep apnea diagnostic agreement, and sleep disordered breathing, apnea, arousal, and leg movement event detection agreement that is substantially equivalent to the predicate device Sleep Profiler (K153412) performance and reference device MICHELE (K112102) performance for all comparisons in all endpoints analyzed respectively. The EnsoSleep performance supports the safety and effectiveness of EnsoSleep when used for the defined indications for use and demonstrates that the device performs as well as the legally marketed predicate Sleep Profiler (K153412) and reference MICHELE Sleep Scoring (K112102).
Key Metrics (Sensitivity, Specificity, PPV, NPV, etc.)
Positive Percent Agreement (PA), Negative Percent Agreement (NA), Overall Percent Agreement (OA), Likelihood ratio (+) and Likelihood ratio (-)
Predicate Device(s): If the device was cleared using the 510(k) pathway, identify the Predicate Device(s) K/DEN number used to claim substantial equivalence and list them here in a comma separated list exactly as they appear in the text. List the primary predicate first in the list.
Advanced Brain Monitoring, Inc. Sleep Profiler (K153412)
Reference Device(s): Identify the Reference Device(s) K/DEN number and list them here in a comma separated list exactly as they appear in the text.
Younes Sleep Technologies MICHELE Sleep Scoring System (K112102)
Predetermined Change Control Plan (PCCP) - All Relevant Information for the subject device only (e.g. presence / absence, what scope was granted / cleared under the PCCP, any restrictions, etc).
Not Found
§ 882.1400 Electroencephalograph.
(a)
Identification. An electroencephalograph is a device used to measure and record the electrical activity of the patient's brain obtained by placing two or more electrodes on the head.(b)
Classification. Class II (performance standards).
0
Image /page/0/Picture/1 description: The image shows the logo for the Department of Health & Human Services - USA. The logo consists of a circular border with the text "DEPARTMENT OF HEALTH & HUMAN SERVICES - USA" written around it. Inside the circle is an emblem that resembles an abstract caduceus, with three stylized human profiles forming the staff and a serpent-like shape winding around them.
Food and Drug Administration 10903 New Hampshire Avenue Document Control Center - WO66-G609 Silver Spring, MD 20993-0002
March 31, 2017
EnsoData, Inc. % Seth Mailhot Partner Michael Best & Friedrich, LLP 601 Pennsylvania Ave, NW Suite 700 South Washington, District of Columbia 20004
Re: K162627
Trade/Device Name: EnsoSleep Regulation Number: 21 CFR 882.1400 Regulation Name: Electroencephalograph Regulatory Class: Class II Product Code: OLZ Dated: March 27, 2017 Received: March 29, 2017
Dear Mr. Mailhot:
We have reviewed your Section 510(k) premarket notification of intent to market the device referenced above and have determined the device is substantially equivalent (for the indications for use stated in the enclosure) to legally marketed predicate devices marketed in interstate commerce prior to May 28, 1976, the enactment date of the Medical Device Amendments, or to devices that have been reclassified in accordance with the provisions of the Federal Food. Drug. and Cosmetic Act (Act) that do not require approval of a premarket approval application (PMA). You may, therefore, market the device, subject to the general controls provisions of the Act. The general controls provisions of the Act include requirements for annual registration, listing of devices, good manufacturing practice, labeling, and prohibitions against misbranding and adulteration. Please note: CDRH does not evaluate information related to contract liability warranties. We remind you, however, that device labeling must be truthful and not misleading.
If your device is classified (see above) into either class II (Special Controls) or class III (PMA), it may be subject to additional controls. Existing major regulations affecting your device can be found in the Code of Federal Regulations, Title 21, Parts 800 to 898. In addition, FDA may publish further announcements concerning your device in the Federal Register.
Please be advised that FDA's issuance of a substantial equivalence determination does not mean that FDA has made a determination that your device complies with other requirements of the Act or any Federal statutes and regulations administered by other Federal agencies. You must comply with all the Act's requirements, including, but not limited to: registration and listing (21 CFR
1
Part 807); labeling (21 CFR Part 801); medical device reporting (reporting of medical devicerelated adverse events) (21 CFR 803); good manufacturing practice requirements as set forth in the quality systems (QS) regulation (21 CFR Part 820); and if applicable, the electronic product radiation control provisions (Sections 531-542 of the Act); 21 CFR 1000-1050.
If you desire specific advice for your device on our labeling regulation (21 CFR Part 801), please contact the Division of Industry and Consumer Education at its toll-free number (800) 638-2041 or (301) 796-7100 or at its Internet address
http://www.fda.gov/MedicalDevices/ResourcesforYou/Industry/default.htm. Also, please note the regulation entitled, "Misbranding by reference to premarket notification" (21 CFR Part 807.97). For questions regarding the reporting of adverse events under the MDR regulation (21 CFR Part 803), please go to
http://www.fda.gov/MedicalDevices/Safety/ReportaProblem/default.htm for the CDRH's Office of Surveillance and Biometrics/Division of Postmarket Surveillance.
You may obtain other general information on your responsibilities under the Act from the Division of Industry and Consumer Education at its toll-free number (800) 638-2041 or (301) 796-7100 or at its Internet address
http://www.fda.gov/MedicalDevices/ResourcesforYou/Industry/default.htm.
Sincerely.
Michael J. Hoffmann -S
for
Carlos L. Peña, PhD, MS Director Division of Neurological and Physical Medicine Devices Office of Device Evaluation Center for Devices and Radiological Health
Enclosure
2
Indications for Use
510(k) Number (if known) K162627
Device Name EnsoSleep
Indications for Use (Describe)
EnsoSleep is intended for use for the diagnostic evaluation by a physician to assess sleep quality and as an aid for the diagnosis of sleep and respiratory related sleep disorders in adults only. EnsoSleep is a software-only medical device to be used under the supervision of a clinician to analyze physiological signals and automatically score sleep study results, including the staging of sleep, detection of arousals, leg movements, and sleep disordered breathing events including obstructive apneas. All automatically scored events are subject to verification by a qualified clinician. Central apneas, mixed apneas, and hypopneas must be manually marked within records.
| Type of Use (Select one or both, as applicable) | |
|---|---|
| ☑ Prescription Use (Part 21 CFR 801 Subpart D) | ☐ Over-The-Counter Use (21 CFR 801 Subpart C) |
CONTINUE ON A SEPARATE PAGE IF NEEDED.
This section applies only to requirements of the Paperwork Reduction Act of 1995.
DO NOT SEND YOUR COMPLETED FORM TO THE PRA STAFF EMAIL ADDRESS BELOW.
The burden time for this collection of information is estimated to average 79 hours per response, including the time to review instructions, search existing data sources, gather and maintain the data needed and complete and review the collection of information. Send comments regarding this burden estimate or any other aspect of this information collection, including suggestions for reducing this burden, to:
Department of Health and Human Services Food and Drug Administration Office of Chief Information Officer Paperwork Reduction Act (PRA) Staff PRAStaff@fda.hhs.gov
"An agency may not conduct or sponsor, and a person is not required to respond to, a collection of information unless it displays a currently valid OMB number."
3
510(k) Summary
| Submitted by: | EnsoData, Inc. |
|---|---|
| Address: | 111 N. Fairchild Street, Suite 240 |
| Madison, WI, 53703 | |
| Telephone: | (608) 509-4704 |
| Contact Name: | Chris Fernandez, Co-founder and CEO |
| Date Submitted: | August 13, 2016 |
| Trade Name: | EnsoSleep |
| Common Name: | Automatic Event Detection Software for Polysomnograph with Electroencephalograph |
| Product Code: | OLZ |
| Regulatory Class: | II (21 C.F.R. 882.1400) |
| Review Panel: | Neurology |
| Predicate Device: | Advanced Brain Monitoring, Inc. Sleep Profiler (K153412) |
| Reference Device: | Younes Sleep Technologies MICHELE Sleep Scoring System (K112102) |
Device Description:
EnsoSleep is a software application that analyzes previously recorded physiological signals obtained during sleep. The EnsoSleep software can analyze any EDF or EDF+ files.
Automated algorithms are applied to the raw signals in order to derive additional signals and interpret the raw and derived signal information. The software automates recognition of:
Sleep Stage Events
- Wake
- Stage N1
- Stage N2
- · Stage N3
- Stage REM
Respiratory Events
- · Sleep disordered breathing (apneas and hypopneas)
- · Apneas detected with airflow signal are classified as obstructive apnea (OSA), and can be edited to be central or mixed appeas
- · Sleep disordered breathing events not detected to be apneas are marked as hypopnea
- · Central apneas, mixed apneas, and hypopneas must be manually marked within records
Arousal Events
· Arousals
- Movement Events
- · Periodic Leg Movements during Sleep (PLMS)
The EnsoSleep software can be used as a stand-alone application for use on Microsoft Windows 7 & 8 operating system platforms. All processing, scoring, and analysis of signal data occurs on the EnsoSleep cloud servers.
Indications for Use:
EnsoSleep is intended for use for the diagnostic evaluation by a physician to assess sleep quality and as an aid for the diagnosis of sleep and respiratory related sleep disorders in adults only. EnsoSleep is a software-only medical device to be used under the supervision of a clinician to analyze physiological signals and automatically score study results, including the staging of arousals, leg movements, and sleep disordered breathing events including obstructive apneas. All automatically scored events are subject to verfication by a qualified clinician. Central apneas, and hypopneas must be manually marked within records.
Determination of Substantial Equivalence:
Summary of Technology:
The EnsoSleep software-only device is similar in inctionality to the Sleep Profiler (K153412) by Advanced Brain Monitoring, Inc. and Younes Sleep Technologies MCHELE Sleep Scoring System (K112102) electroencephalograph analysis software programs. The EnsoSleep device is similar with respect to indications for use and physical characteristics to the predicate device and
4
reference device in terms of 510(k) substantial equivalency.
Based on analysis and comparison of technological characteristics and features between EnsoSleep and reference device, EnsoSleep is determined to have the same technological characteristics as the predicate and reference devices, and does not raise different questions of safety or efficacy as demonstrated by the device design. EnsoSleep uses the same fundamental technology as the legally marketed predicate device; automated algorithms are applied to raw signals in order to derive additional signals and interpret raw and derived signal information. Each of the EnsoSleep, Sleep Profiler (K153412), and MCHELE Sleep Scoring System (K112102) devices include as features the ability for full disclosure recording of derived signals and automated analyses to be visually inspected and edited prior to the results being integrated into one of several sleep study report data formats. Each of the EnsoSleep, Sleep Profiler (K153412), and MCHELE Sleep Scoring System (K112102) devices base the automatic scoring of physiological events on the American Academy of Sleep Medicines, definitions, and procedures. Additionally, the EnsoSleep predicate device Sleep Profiler (K153412, K13007, K120450), in an earlier cleared version of the device, utilized MCHELE Sleep Scoring (K112102) as its own predicate device for establishing substantial equivalence validation testing comparisons.
Both the predicate Sleep Profiler (K153412) and EnsoSleep devices can be used as a stand-alone software application with a user interface delivered on the Microsoft Windows 7 or 8 operating system platforms, and automatically reject periods of poor EEG signal quality. The following features were not included in the EnsoSleep application as they were deemed unnecessary based on end-user feedback and/or the fact that they are not requirement AASM guidelines: detection of heart rate, head position, snoring levels, head movements, and respiratory event related apneas (RERAs); disease management comments; two-night reports. The chosen EnsoSleep design and implementation features scoing, efficiency, reliability, and security improvements over Sleep Profiler (K153412) and MICHELE Sleep Scoring System (K112102). EnsoSleep provides more comprehensive coverage of the event types specified for scoring by the AASM Manual for Scoring and Associated Events recommendations than the Sleep Profiler (K153412) device by automatically scoring leg movement events in addition to the sleep disordered breathing, obstructive apnea, arousal, and sleep staging functionality supported by both the predicate and reference devices. Both the Sleep Profiler (K153412) predicate device and EnsoSleep classify each apnea as obstructive, each non-apnea sleep disordered breathing event as a hypopnea (excluding RERAs), and enable apneas to be edited and manually marked to be central or mixed apneas within records.
EnsoSleep differs from Sleep Profiler (K153412) and MICHELE Sleep Scoring System (K112102) in that EnsoSleep automates the initiation, upload, scoring, and download of studies, enabling a user experience that is optimized for efficiency and fast analysis prior to the required user over-read of EnsoSleep scring, Futhermore, the high-performance specifications of the EnsoSleep Processing Platform enables throughput efficiency and scalability, as the speed by which a study processes is dependent on the number of distributed servers, parallel computer processor(s), and the amount of RAM. EnsoSleep improves network protocols with automatic connection-recovery that are robust to poor or interrupted network conditions. Finally, EnsoSleep provides cybersecurity improvements over the predicate Sleep Profiler (K153412) and MCHELE Sleep Scoring System (K112102) with verified authentication, authorization, and access controls and checksum controls including end-to-end encryption; secure software distribution mechanisms and controls, intrusion systems and vulnerability scanning; and other network, systems, and database controls.
Summary of Non-Clinical Tests:
Support for the substantial equivalence of EnsoSleep was provided by risk management and software testing. Both EnsoSleep and Sleep Profiler (K153412) conduct and document verification, and performance testing as recommended by FDA's Guidance for Industry and FDA Staff, "Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices". The EnsoSleep software has been thorough verification of specifications. One or more verification tests are provided for each requirement specified with detailed protocols, objective passfial criteria, and clearly documented test executions with results. Detailed plans and protocols were developed prospectively for system level, performance, and usability validation testing was conducted with qualified clinical and non-clinical users, with objective passffall criteria, ability to provide comments, and with all testing reports and results documented for review.
Summary of Clinical Tests:
Substantial equivalence was also established through a testing protocol that used clinical polysomnography (PSG) data to evaluate the performance of EnsoSleep. Clinical performance testing was completed by evaluating EnsoSleep device performance using a cross-sectional experimental design on a representative N=72 subject sample of retrospective clinical PSG data. First, the intended use population, study population, conditions of interest, designated comparative reference, designated comparative benchmarks, and experimental endpoints were defined. Upon applying predeined selection controls, a statistically representative sample of the defined intended use and user population sample consisting of N=22 archived collection of retrospective diagnostic clinical PSG data collected from an AASM Accredity. Same as the predicate and reference devices respectively, the study population was then sent to a clinical testing laboratory where each PSG was manually scored by three (3) independent registered sleep technologists (RPSGT) that met all acquisition, sooring-bind, and rater controls. A designated comparative reference was constructed using 2/3 Majority Scoring to evaluate the EnsoSleep device performance versus the predicate Sleep Profiler and MICHELE device performance benchmarks and acceptance criteria of positive percent agreement (PA), negative percent agreement (NA), and overall percent (OA) were predefined competitively based on analysis of sleep staging event detection and diagnostic agreement performance reported in the predicate device 510(k) documentation. The predicate device did not report dinical testing results for shing, arousal, and leg movement event detection agreement performance, and therefore the reference device clinical performance and 510(k) documentation were used to facilitate a valid comparison. EnsoSleep device performance was evaluated using the definental design, statistical methodology, and controls, across the following three (3) experimental endpoints:
-
- Endpoint 1: As EnsoSleep is intended to assist clinicians with the assessment of sleep scoring must be validated. For Endpoint 1, EnsoData evaluated a performance goal comparing the predicated ovice Sleep Profiler (K153412) PA, NA, and OA sleep staging performance, and the bootstrapped point estimate of EnsoSleep sleep staging PA, NA, and OA versus a 2/3 Majority Scoring reference.
5
-
- Endpoint 2: As EnsoSleep is intended to assist clinicians with the scoring sleep disordered breathing events used in diagnostic evaluation, device performance for diagnosing sleep apnea must be validated. For Endpoint 2, EnsoData evaluated a performance goal comparing the predicate device Sleep Profiler (K153412) PA. NA, and OA diagnostic agreement performance, and the bootstrapped point estimate of median performance for EnsoSleep diagnostic agreement PA, NA, and OA versus a 2/3 Majority Scoring reference.
-
- Endpoint 3: As EnsoSleep is intended to analyze physiological signals and automatically score sleep study results, including detection of sleep disordered breathing events, arousal events, and leg movement events, device performance for detecting each event type must be validated. For Endpoint 3, EnsoData evaluated a performance goal comparing the reference device MICHELE Sleep Scoring System (K112102) PA, NA, and OA event detection performance, and the bootstrapped point estimate of median performance for EnsoSleep event detection PA, NA, and OA versus a 2/3 Majority Scoring reference.
The final experimental results and statistical and ont-on-NA, OA performance with 95% percentile bootstrap confidence intervals (R=1000 resamples) were calculated by overall-pochs versus 2/3 Majority Scoring in event detection experiments evaluating Wake, N1, N2, N3, REM, SDB, Apneal, and Leg Movement event detection performance respectively. Furthermore. EnsoSleep device diagnostic agreement was evaluated versus 2/3 Majority on both mild and moderate sleep apnea diagnostic thresholds by computing the overall and REM-only apneal. bootstrapped point-estimates for median PA, NA, OA, performance with 95% percentile bootstrap confidence intervals (Re1000 resamples) and likelihood ratio pairs were computed in diagnosic agreement experiments evaluating overall-mild AHI, REM-mid AHI, and REM-moderate AHI. The final experimental results are summarized for each endpoint in Table 1, Table 2, and Table 3 below.
Table 1 shows the EnsoSleep and Sleep Profiler (K153412) clinical performance results compared for Endboint 1, sleep staging. For all sleep staging event types evaluated in EnsoSleep PA, NA, and OA performance was observed to show no statistically significant differences or was observed to be significantly greater in all comparisons relative to the predicate device. There were no cases where predicate device Wake, N1, N2, N3, REM, or Total Staging performance was statistically greater than EnsoSleep event detection performance (i.e. the precicate device point estimate for PA. NA, and OA were never higher than EnsoSleep PA, NA, and OA 95% CI upper bounds). The results confirm EnsoSleep achieves clinical performance for sleep staging positive, negative, and overall agreement that is substantially equivalent to the Sleep Profiler positive, and overall agreement across all sleep stages.
| Table 1: Sleep Staging Clinical
Performance Comparisons | | | Overall-Epochs EnsoSleep
vs 2/3 Majority Sleep Staging Performance | | | Overall-Epochs Sleep Profiler (K153412)
vs 2/3 Majority Sleep Staging Performance | | | |
|------------------------------------------------------------|-------|----------------------------------|-------------------------------------------------------------------------------------------------------------------------------------------|----------------------------------|---------------------------------------------------------------------------------|--------------------------------------------------------------------------------------|--|--|--|
| | | (N=72 subjects,
59719 epochs) | Bootstrapped point-estimate of median
Percent Agreement (%) with 95%
percentile bootstrap confidence interval
(R=1000 resamples) | (N=43 subjects,
31361 epochs) | Point-estimate of Percent Agreement (%) | | | | |
| | | Total Epochs | Positive Agreement
(PA) Negative Agreement
(NA) Overall Agreement
(OA) | Total Epochs | Positive Agreement
(PA) Negative Agreement
(NA) Overall Agreement
(OA) | | | | |
| | Wake | 17459 | 86%
(82%, 88%) 97%
(95%, 98%) 94%
(92%, 95%) | 7424 | 73% 94% 89% | | | | |
| Overall-epochs
assigned by 2/3
Majority Scoring | N1 | 3293 | 41%
(33%, 48%) 94%
(93%, 96%) 91%
(90%, 93%) | 1752 | 25% 93% 89% | | | | |
| | N2 | 26839 | 77%
(73%, 81%) 87%
(85%, 90%) 83%
(80%, 85%) | 12582 | 77% 84% 81% | | | | |
| | N3 | 5587 | 81%
(74%, 88%) 93%
(91%, 95%) 92%
(90%, 94%) | 4704 | 76% 94% 91% | | | | |
| | REM | 6541 | 79%
(72%, 84%) 99%
(98%, 99%) 96%
(96%, 97%) | 3749 | 74% 97% 95% | | | | |
| | Total | 59719 | 78%
(77%, 80%) 95%
(94%, 95%) 91%
(91%, 92%) | 31361 | 73% 93% 87% | | | | |
| | None | 1432 | - - - | 1150 | - - - | | | | |
Table 2 shows the EnsoSleep and Sleep Profiler (K153412) dinical performance results compared for Endposition agreement. For all diagnostic agreements evaluated in Endpoint 2, mild-REM, moderate-overall, and moderate-REM, EnsoSleep PA and NA performance were observed to show no statistically significant differences compared to predicate device performance (e.g. the two-sided 95% percentle bounds contained the predicate device point estimates for PA and NA in all cases). The only statistically significant difference observed was for overall-mild PA, with a 2% difference in the EnsoSleep Cl upper bound and Sleep Profiler point estimate (91% (82%,98%) vs. 100%) and no statistically significant differences in overall-mild NA. Furthermore, the point estimate of EnsoSleep PA, NA, and OA performance exceeded, were equivalent to, or were within 10% of the predicate device PA and NA point estimates for all comparisons. The EnsoSleep positive likelihood ratios were observed to be above 3.5 for overall/REM-mild and above 5.0 for overall/REM-moderate in all diagnostic agreement experiments, similarly exceeding the performance goal targeted by Sleep Profiler (K153412) in the predicate device 510(k) documentation. The resoSleep achieves clinical performance for positive sleep apnea diagnostic agreement that is substantially equivalent to the Sleep Profiler positive and negative agreement across all mild-REM, moderate-overall, and moderate-REM comparisons.
6
| | Per-Patient EnsoSleep vs 2/3 Majority
Sleep Apnea Diagnostic Agreement | | Per-Patient Sleep Profiler (K153412) vs 2/3
Majority
Sleep Apnea Diagnostic Agreement | | | | | |
|------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------|------------------------|---------------------------------------------------------------------------------------------|----------------------------|----------------------------|-----------------------------|--------------------------------|---------------------------------|
| Table 2: Sleep Apnea Diagnostic Agreement
Clinical Performance Comparisons
EnsoSleep bootstrapped point-estimate of
median Percent Agreement (%) with 95%
percentile bootstrap confidence interval
(R=1000 resamples), Sleep Profiler
point-estimate of Percent Agreement (%), and
likelihood ratio pairs | EnsoSleep
AHI >= 5 | EnsoSleep
AHI >= 15 | EnsoSleep REM
AHI >= 5 | EnsoSleep REM
AHI >= 15 | Sleep Profiler
AHI >= 5 | Sleep Profiler
AHI >= 15 | Sleep Profiler REM
AHI >= 5 | Sleep Profiler REM
AHI >= 15 |
| Sample size (N) | 72 | 72 | 72 | 72 | 60 | 60 | 40 | 40 |
| Positive Agreement (PA) | 91%
(82%, 98%) | 95%
(83%, 100%) | 83%
(72%, 94%) | 79%
(56%, 94%) | 100% | 100% | 84% | 73% |
| Negative Agreement (NA) | 76%
(61%, 90%) | 98%
(94%, 100%) | 89%
(79%, 97%) | 96%
(90%, 100%) | 85% | 97% | 90% | 96% |
| Overall Agreement (OA) | 85%
(77%, 92%) | 97%
(93%, 100%) | 86%
(79%, 93%) | 92%
(85%, 97%) | N/A | N/A | N/A | N/A |
| Likelihood ratio (+) | 3.76 | 52.25 | 7.71 | 22.00 | 6.67 | 9.50 | 8.84 | 18.33 |
| Liklihood ratio (-) | 0.12 | 0.05 | 0.19 | 0.22 | 0.00 | 0.00 | 0.00 | 0.28 |
Table 3 shows the EnsoSleep and MICHELE Sleep Scoring (K112102) clinical performance results compared for Endpoint 3, event detection agreement. The MCHELE Sleep Scoring performance testing did not calculate OA for individual event types in the same way as EnsoSleep, and as such only PA and NA comparisons were made to avoid biased performance evaluation. For all event detection experiments including SDB, OSA, Arousal, and Leg Movement event types, the point-estimates of EnsoSleep PA and NA event detection performance exceeded, were equivalent to, or were within 10% of the refermance, with statistically significant differences observed in a minority of cases. On the basis that EnsoSleep met or exceeded objective PA and NA performance goals for these event types in all comparisons, and the reference device did not provide information to adequately compare the statistical significance of results (i.e. two-sided confidence intervals for point-estimates of agreement were not reported), EnsoSleep is considered substantially equivalent to the reference device SDB, Arousal, and Leg Movement event detection performance.
| Table 3: Event Detection Clinical
Performance Comparisons | | | | Overall-Epochs MICHELE (K112102)
vs 2/3 Majority Event Detection Performance | | | | | |
|----------------------------------------------------------------------|----------------------------------------------------------------------------------------------------------------------------------|-------------------------------|----------------------------------------------------------------------------------------------------------------------------------|---------------------------------------------------------------------------------|----------------|-------------------------|-------------------------|------------------------|-------|
| Overall-Epochs EnsoSleep vs 2/3 Majority Event Detection Performance | | | | | | | | | |
| (N=72 subjects, 59719 epochs) | Bootstrapped point-estimate of median Percent Agreement (%) with 95% percentile bootstrap confidence interval (R=1000 resamples) | (N=30 subjects, 24987 epochs) | Weighted average of Positive and Negative Percent Agreement (%) point-estimates and Overall Percent Agreement (%) point-estimate | | | | | | |
| | Total Epochs | Positive Agreement (PA) | Negative Agreement (NA) | Overall Agreement (OA) | Total Epochs | Positive Agreement (PA) | Negative Agreement (NA) | Overall Agreement (OA) | |
| Overall-epochs assigned by 2/3 Majority Scoring | Sleep Disordered Breathing Events (apneas and hypopneas) | 4705 | 67% (58%, 75%) | 93% (92%, 94%) | 91% (90%, 92%) | 2439 | 75.5% | 98.1% | 93.0% |
| | Apnea Events | 1690 | 56% (41%, 70%) | 96% (96%, 97%) | 95% (95%, 96%) | 750 | N/A | N/A | N/A |
| | Obstructive Apnea Events | 1066 | 53% (35%, 71%) | 97% (96%, 97%) | 96% (95%, 97%) | 359 | 55.9% | 99.3% | N/A |
| | Arousal Events | 7686 | 66% (61%, 71%) | 90% (88%, 91%) | 87% (85%, 88%) | 2278 | 60.0% | 94.1% | 89.9% |
| | Leg Movement Events | 5796 | 71% (60%, 80%) | 90% (89%, 92%) | 89% (87%, 90%) | 1714 | 78.4% | 97.6% | 95.7% |
In summary, performance test results demonstrated that EnsoSleep staging event detection, sleep apnea diagnostic agreement, and sleep disordered breathing, apnea, arousal, and leg movement event detection agreement that is substantially equivalent to the predicate device Sleep Profiler (K153412) performance and reference device MICHELE (K153412) performance for all comparisons in all endpoints analyzed respectively. The EnsoSleep performance the safety and effectiveness of EnsoSleep when used for the defined indications for use and demonstrates that the device performs as well as the legally marketed predicate Sleep Profiler (K153412) and reference MICHELE Sleep Scoring (K112102).
Conclusion:
Nor-Clinical and Clinical verification, and performance testing was conducted in accordance with FDA guidance recommendations to confirm the device design met all specifications, user needs, and was acceptable to qualified clinical and non-clinical users. EnsoSleep has passed all of the aforementioned Verification and Validation tests and provided Clinical Performance testing results with a library clinical dataset in order to demonstrate safety or effectiveness. It is therefore concluded that EnsoSleep is substantially equivalent to the predicate device.