(192 days)
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.
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.
Here's a breakdown of the acceptance criteria and study details for the EnsoSleep device, based on the provided FDA 510(k) summary:
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria were established competitively based on the performance reported in the predicate device (Advanced Brain Monitoring, Inc. Sleep Profiler, K153412) and the reference device (Younes Sleep Technologies MICHELE Sleep Scoring System, K112102) 510(k) documentation. The study aimed to demonstrate that EnsoSleep's performance was statistically similar to or greater than these benchmarks.
Endpoint 1: Sleep Staging Performance
| Sleep Stage (overall-epochs) | EnsoSleep Performance (Bootstrapped Point Estimate & 95% CI) | Predicate Device (Sleep Profiler K153412) Performance (Point Estimate) |
|---|---|---|
| Wake | PA: 86% (82%, 88%), NA: 97% (95%, 98%), OA: 94% (92%, 95%) | PA: 73%, NA: 94%, OA: 89% |
| N1 | PA: 41% (33%, 48%), NA: 94% (93%, 96%), OA: 91% (90%, 93%) | PA: 25%, NA: 93%, OA: 89% |
| N2 | PA: 77% (73%, 81%), NA: 87% (85%, 90%), OA: 83% (80%, 85%) | PA: 77%, NA: 84%, OA: 81% |
| N3 | PA: 81% (74%, 88%), NA: 93% (91%, 95%), OA: 92% (90%, 94%) | PA: 76%, NA: 94%, OA: 91% |
| REM | PA: 79% (72%, 84%), NA: 99% (98%, 99%), OA: 96% (96%, 97%) | PA: 74%, NA: 97%, OA: 95% |
| Total (Overall Epochs) | PA: 78% (77%, 80%), NA: 95% (94%, 95%), OA: 91% (91%, 92%) | PA: 73%, NA: 93%, OA: 87% |
Conclusion on Endpoint 1: EnsoSleep showed no statistically significant differences or was significantly greater than the predicate device in all comparisons for sleep staging performance.
Endpoint 2: Sleep Apnea Diagnostic Agreement Performance
| Diagnostic Agreement (Per-Patient) | EnsoSleep Performance (Bootstrapped Point Estimate & 95% CI) | Predicate Device (Sleep Profiler K153412) Performance (Point Estimate) |
|---|---|---|
| AHI >= 5 (overall-mild) | PA: 91% (82%, 98%), NA: 76% (61%, 90%), OA: 85% (77%, 92%) | PA: 100%, NA: 85% |
| AHI >= 15 (overall-moderate) | PA: 95% (83%, 100%), NA: 98% (94%, 100%), OA: 97% (93%, 100%) | PA: 100%, NA: 97% |
| REM AHI >= 5 (REM-mild) | PA: 83% (72%, 94%), NA: 89% (79%, 97%), OA: 86% (79%, 93%) | PA: 84%, NA: 90% |
| REM AHI >= 15 (REM-moderate) | PA: 79% (56%, 94%), NA: 96% (90%, 100%), OA: 92% (85%, 97%) | PA: 73%, NA: 96% |
Conclusion on Endpoint 2: EnsoSleep showed no statistically significant differences in PA and NA compared to the predicate device, with one exception for overall-mild PA (a 2% difference in EnsoSleep's CI upper bound vs. predicate point estimate). EnsoSleep's point estimates for PA, NA, and OA exceeded, were equivalent to, or were within 10% of the predicate device's.
Endpoint 3: Event Detection Performance
| Event Type (Overall-Epochs) | EnsoSleep Performance (Bootstrapped Point Estimate & 95% CI) | Reference Device (MICHELE Sleep Scoring K112102) Performance (Point Estimate) |
|---|---|---|
| SDB | PA: 67% (58%, 75%), NA: 93% (92%, 94%), OA: 91% (90%, 92%) | PA: 75.5%, NA: 98.1%, OA: 93.0% |
| Apnea | PA: 56% (41%, 70%), NA: 96% (96%, 97%), OA: 95% (95%, 96%) | N/A (did not provide PA/NA) |
| Obstructive Apnea | PA: 53% (35%, 71%), NA: 97% (96%, 97%), OA: 96% (95%, 97%) | PA: 55.9%, NA: 99.3% |
| Arousal | PA: 66% (61%, 71%), NA: 90% (88%, 91%), OA: 87% (85%, 88%) | PA: 60.0%, NA: 94.1%, OA: 89.9% |
| Leg Movement | PA: 71% (60%, 80%), NA: 90% (89%, 92%), OA: 89% (87%, 90%) | PA: 78.4%, NA: 97.6%, OA: 95.7% |
Conclusion on Endpoint 3: EnsoSleep's point-estimates for PA and NA event detection performance exceeded, were equivalent to, or were within 10% of the reference device, with statistically significant differences observed in a minority of cases.
2. Sample Size for Test Set and Data Provenance
- Sample Size (Test Set): N=72 subjects (59719 epochs for epoch-based analyses, 72 for per-patient analyses).
- Data Provenance: Retrospective clinical PSG data. The data constituted an "archived collection of retrospective diagnostic clinical PSG data collected from an AASM Accredited" facility. The country of origin is not explicitly stated but implied to be the US given the FDA submission.
3. Number of Experts for Ground Truth and Qualifications
- Number of Experts: Three (3) independent registered sleep technologists (RPSGT).
- Qualifications: Registered sleep technologists (RPSGT) who met "all acquisition, scoring-bind, and rater controls." Specific years of experience are not mentioned.
4. Adjudication Method for Test Set
- Adjudication Method: 2/3 Majority Scoring was used to establish the designated comparative reference (ground truth).
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- No MRMC comparative effectiveness study was done to evaluate human readers with and without AI assistance. The study described is a standalone performance evaluation of the AI algorithm against expert consensus.
6. Standalone (Algorithm Only) Performance Study
- Yes, a standalone study was performed. The entire clinical performance testing described evaluates the EnsoSleep algorithm's performance (without human-in-the-loop) against a human expert consensus ground truth.
7. Type of Ground Truth Used
- Type of Ground Truth: Expert consensus (2/3 Majority Scoring by three independent RPSGTs).
8. Sample Size for Training Set
- The document does not specify the sample size for the training set. It only describes the validation/test set.
9. How Ground Truth for Training Set Was Established
- The document does not provide information on how the ground truth for the training set was established. It focuses solely on the clinical performance testing (validation/test set).
§ 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).