Search Results
Found 1 results
510(k) Data Aggregation
(149 days)
Neumetry Medical Inc
SomnoMetry 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. SomnoMetry is a software-only medical device to be used to analyze physiological signals and automatically score sleep study results, including the staging of sleep, AHI, and detection of sleepdisordered breathing events including obstructive apneas. It is intended to be used under the supervision of a clinician in a clinical environment. All automatically scored events are subject to verification by a qualified clinician.
The SomnoMetry is an Artificial Intelligent/Machine Learning (AI/ML)-enabled Software as a Medical Device (SaMD) that automatically scores sleep study results by analyzing polysomnography (PSG) signals recorded during sleep studies. It is intended to be used under the supervision of a clinician in clinical environments to aid in the diagnosis of sleep and respiratory related sleep disorders.
All scored events that are analyzed, displayed, and summarized can be manually marked or edited by a qualified clinician during review and verification.
SomnoMetry consists of:
- A web Application Programing Interface (API) to allow authenticated users to upload . PSG files to SomnoMetry Platform
- A database to store the input, intermedium output, final output, and associated data ●
- A database API to access the database and store/retrieve the output ●
- A dashboard to display, retrieve, manage, edit, verify, and summarize the output
- An AI/ML Engine using AI/ML algorithms/approaches to analyze PSG data ●
- A reporting API to generate sleep reports
Here's an analysis of the acceptance criteria and study that proves the device meets them, based on the provided text:
Device: SomnoMetry (Neumetry Medical Inc.)
Regulatory Class: Class II (Product Code: OLZ)
Intended Use: Diagnostic evaluation to assess sleep quality and aid in diagnosing sleep and respiratory-related sleep disorders in adults. Automatically scores sleep study results (sleep staging, AHI, obstructive apneas). Used under clinician supervision; all automatically scored events are subject to verification.
1. Table of Acceptance Criteria and Reported Device Performance
The document doesn't explicitly define 'acceptance criteria' in a numerical table for the SomnoMetry device against specific thresholds (e.g., "accuracy must be > 90%"). Instead, it frames the performance evaluation as demonstrating non-inferiority and substantial equivalence to a predicate device (EnsoSleep K162627) based on clinical performance metrics. The implicit acceptance criteria are that the SomnoMetry's performance for sleep staging and sleep apnea diagnosis is statistically equivalent to, or better than, the predicate device.
Table 1: Implicit Acceptance Criteria and SomnoMetry's Reported Performance (vs. Predicate)
| Metric/Endpoint | Acceptance Criterion (Implicit: Non-inferiority/Substantial Equivalence to Predicate) | SomnoMetry Performance (Clinical Study Results) | Predicate Device Performance (Clinical Study Results) |
|---|---|---|---|
| Endpoint 1: Sleep Staging Performance | |||
| Wake (W) | Non-inferior to AASM gold standard (manual scoring) | True Label W: Predicted W = 92.7% (91.8, 93.6)* | Not directly comparable in this table; stated as "substantially equivalent to the predicate device." |
| Stage 1 (N1) | Non-inferior to AASM gold standard (manual scoring) | True Label N1: Predicted N1 = 47.1% (46.1, 48.8)* | Not directly comparable in this table; stated as "substantially equivalent to the predicate device." |
| Stage 2 (N2) | Non-inferior to AASM gold standard (manual scoring) | True Label N2: Predicted N2 = 88.3% (87.4, 89.1)* | Not directly comparable in this table; stated as "substantially equivalent to the predicate device." |
| Stage 3 (N3) | Non-inferior to AASM gold standard (manual scoring) | True Label N3: Predicted N3 = 80.8% (79.8, 81.7)* | Not directly comparable in this table; stated as "substantially equivalent to the predicate device." |
| REM (R) | Non-inferior to AASM gold standard (manual scoring) | True Label R: Predicted R = 94.5% (93.5, 95.5)* | Not directly comparable in this table; stated as "substantially equivalent to the predicate device." |
| *Note on Sleep Staging: | The confusion matrix shows how well the algorithm predicts each sleep stage given the true stage. Higher percentages on the diagonal (true W to predicted W, etc.) indicate better performance. | ||
| Endpoint 2: Sleep Apnea Diagnostic Agreement (AHI Thresholds) | |||
| Positive Agreement (PA) | No statistically significant differences from predicate device | All: AHI > 5: 90.6%; AHI ≥ 15: 89.1%; AHI ≥ 30: 83.3% | |
| REM: AHI ≥ 5: 85.6%; AHI ≥ 15: 80.0%; AHI ≥ 30: 78.8% | All: AHI > 5: 91%; AHI ≥ 15: 95%; AHI ≥ 30: N/A | ||
| REM: AHI > 5: 83%; AHI ≥ 15: 79%; AHI ≥ 30: N/A | |||
| Negative Agreement (NA) | No statistically significant differences from predicate device | All: AHI > 5: 92.2%; AHI ≥ 15: 94.9%; AHI ≥ 30: 97.5% | |
| REM: AHI ≥ 5: 94.7%; AHI ≥ 15: 94.7%; AHI ≥ 30: 95.6% | All: AHI > 5: 76%; AHI ≥ 15: 98%; AHI ≥ 30: N/A | ||
| REM: AHI > 5: 89%; AHI ≥ 15: 96%; AHI ≥ 30: N/A | |||
| Overall Agreement (OA) | No statistically significant differences from predicate device | All: AHI > 5: 91.2%; AHI ≥ 15: 92.8%; AHI ≥ 30: 95.6% | |
| REM: AHI ≥ 5: 88.9%; AHI ≥ 15: 88.9%; AHI ≥ 30: 92.4% | All: AHI > 5: 85%; AHI ≥ 15: 97%; AHI ≥ 30: N/A | ||
| REM: AHI > 5: 86%; AHI ≥ 15: 92%; AHI ≥ 30: N/A | |||
| Likelihood Ratio (+) | No statistically significant differences from predicate device | All: AHI > 5: 11.62; AHI ≥ 15: 17.47; AHI ≥ 30: 33.32 | |
| REM: AHI ≥ 5: 16.15; AHI ≥ 15: 15.09; AHI ≥ 30: 17.91 | All: AHI > 5: 3.76; AHI ≥ 15: 52.25; AHI ≥ 30: N/A | ||
| REM: AHI > 5: 7.71; AHI ≥ 15: 22.0; AHI ≥ 30: N/A | |||
| Likelihood Ratio (-) | No statistically significant differences from predicate device | All: AHI > 5: 0.10; AHI ≥ 15: 0.11; AHI ≥ 30: 0.17 | |
| REM: AHI ≥ 5: 0.15; AHI ≥ 15: 0.21; AHI ≥ 30: 0.22 | All: AHI > 5: 0.12; AHI ≥ 15: 0.05; AHI ≥ 30: N/A 5: 0.19; AHI ≥ 15: 0.22; AHI ≥ 30: N/A |
2. Sample Size and Data Provenance for Test Set
- Sample Size: N = 201 adult subjects.
- Data Provenance: Retrospective clinical data obtained from 2 AASM accredited Sleep Testing Facilities in California, USA.
- The data was verified to meet specified disease spectrum, medical condition, medication, and demographic requirements.
- Randomized sampling with proportionate allocation across each sleep apnea disease severity quantile (normative, mild, moderate, and severe sleep apnea) and sleep cycles was used.
- Age of subjects ranged from 20 to 84 years.
- No race/ethnicity information was collected.
3. Number of Experts and Qualifications for Ground Truth
The document explicitly states the ground truth was established by "manually scored PSG data" which aligns with the "AASM gold standard". While it doesn't specify the number of experts, it implies these were the standard scores provided by the AASM accredited facilities. The qualifications are implicitly that they are "qualified clinicians" following "American Academy of Sleep Medicine scoring manual and guidelines."
4. Adjudication Method for Test Set
The document does not describe a formal expert adjudication method (e.g., 2+1, 3+1). The ground truth is stated as "AASM gold standard of manually scored PSG data," which suggests a single, accepted manual score per PSG study generated by the sleep facility.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No, a MRMC comparative effectiveness study was not conducted. The study evaluated the standalone performance of the AI algorithm against the AASM gold standard and then compared this algorithm performance to the predicate device's algorithm performance. It did not assess how human readers improve with AI assistance versus without. The device's intended use note ("All automatically scored events are subject to verification by a qualified clinician") implies a human-in-the-loop workflow, but the study itself wasn't designed to measure the impact of AI assistance on human reader performance.
6. Standalone (Algorithm Only) Performance Study
Yes, a standalone (algorithm only) performance study was performed. The "retrospective clinical performance testing" evaluated the SomnoMetry AI/ML algorithms directly, comparing their output to the AASM gold standard. The results presented in Table 2 (Confusion Matrix for Sleep Staging) and Table 3 (Sleep Apnea Diagnostic Agreement) are for the algorithm's performance.
7. Type of Ground Truth Used
The type of ground truth used was expert consensus / AASM (American Academy of Sleep Medicine) gold standard based on "manually scored PSG data." This is considered the clinical standard for sleep study interpretation.
8. Sample Size for the Training Set
The document does not specify the sample size for the training set. It only describes the test set (N=201).
9. How Ground Truth for the Training Set Was Established
The document does not describe how the ground truth for the training set was established. It focuses solely on the clinical evaluation (test set) and states that data was obtained from AASM accredited facilities. It can be inferred that similar "manually scored PSG data" would have been used for training, given the expertise required for such scoring and the device's reliance on AASM guidelines.
Ask a specific question about this device
Page 1 of 1