Search Results
Found 1 results
510(k) Data Aggregation
(54 days)
COMPASS F10 SYSTEM
The intended use of the Compass F10 system is to record physiological signals during sleep, scan the signals for abnormalities and represent the count of abnormal events in a form of a summary report. The results of the scan may be manually overwritten or corrected by the physician. The device is intended for use as a screening device to determine the need for clinical diagnosis and evaluation by polysomnography based on the patient's count of abnormal events. It is not intended for any diagnosis. It is not intended to be a monitor.
The Compass F10 device is intended to be used for adult and pediatric patients.
The Compass F10 system is an ambulatory recording system. It includes a recording device, a signal adapter, strap system for attaching of recording device to a patient, an USB cable for data download and the Compass application.
The Compass F10 device is a pocket size battery powered digital recorder that incorporates electronics to record and store one night of physiological parameters. It has one respiratory channel for measurement of nasal pressure/airflow and a built-in body position and actigraph sensor for measurement of body position and movement. It also has an optional oximeter input to measure degree of oxygen saturation of the blood.
The Compass application provides the means to prepare the device for recording, download the recorded data, viewing and analyzing the recorded data on a PC.
Here's a detailed breakdown of the acceptance criteria and the study proving the device meets them, based on the provided text:
Acceptance Criteria and Device Performance Study for Compass F10 System
1. Table of Acceptance Criteria and Reported Device Performance
The document does not explicitly state pre-defined quantitative acceptance criteria in a table format. However, it presents the performance metrics of the Compass F10 system and compares them against a qualitative assessment of "good estimate" and "performs very well." It also implicitly uses the performance of the predicate device (ApLab) as a benchmark.
For the purpose of this analysis, we infer the de facto acceptance criteria to be achieving correlation and agreement with manual PSG scoring, and performing comparably or better than the predicate ApLab system.
| Performance Metric | Compass F10 + Oximeter Performance | Compass F10 (without Oximeter) Performance | Implied Acceptance Criterion (based on "good estimate" and predicate comparison) |
|---|---|---|---|
| Sensitivity | 100% | 96.8% | High sensitivity (e.g., comparable to or better than ApLab's 89%) for accurately identifying abnormal events. |
| Specificity | 87% | 82.6% | Reasonable specificity to minimize false positives. Not explicitly quantified but implied by "good estimate." |
| PPV (Positive Predictive Value) | 91.2% | 88.2% | High PPV to ensure that positive results are likely true positives. Not explicitly quantified but implied by "good estimate." |
| NPV (Negative Predictive Value) | 100% | 95.0% | High NPV to ensure that negative results are likely true negatives. Not explicitly quantified but implied by "good estimate." |
| Correlation | 96.98% | 88.59% | High correlation with manually scored AHI values (e.g., above 85-90%). |
| Bland & Altman 95% Confidence Interval | -6.3 to 6.1 (average -0.1) | -14.3 to 10.6 (average -1.8) | Narrow 95% confidence interval with an average close to zero, indicating good agreement with manual scoring. |
| Comparison to Predicate ApLab Sensitivity | 100% (Compass F10 + oximeter) | 96.8% (Compass F10 without oximeter) | Performance (at least sensitivity) for identifying abnormal events should be comparable to or better than ApLab (89%). |
| Overall Assessment | "performs very well in all cases." | "performs very well in all cases." | The device should provide a "good estimate" of AHI values and effectively distinguish between "Normals and potential OSA's patient." |
2. Sample Size Used for the Test Set and Data Provenance
- Test Set Sample Size: Not explicitly stated. The document mentions "The validation data used was recorded with Embla N7000 using full polysomnography (PSG)." It does not specify the number of patients or recordings in this validation dataset.
- Data Provenance: The data was recorded using an Embla N7000, which is mentioned as a predicate device from Medcare Flaga. The country of origin for the data is not specified. It is implied to be retrospective as the recordings were already performed using the Embla N7000 and then hand-scored.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications
- Number of Experts: "The recordings were all hand scored by the same RPSG technician." This indicates one expert.
- Qualifications of Expert: "RPSG technician." (Registered Polysomnographic Technician)
4. Adjudication Method for the Test Set
- Adjudication Method: None, for the initial ground truth. The ground truth was established by a single RPSG technician. The Compass application "results of the scan may be manually overwritten or corrected by the physician," but this is post-analysis by the device, not part of establishing the ground truth for its own validation.
5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done
- No, an MRMC comparative effectiveness study was not done. The study compared the Compass automatic scoring to a single manual scoring (ground truth). There is no mention of multiple human readers evaluating cases with and without AI assistance or any analysis of the effect size of AI on human reader performance.
6. If a Standalone (Algorithm Only Without Human-in-the-Loop Performance) Was Done
- Yes, a standalone study was done. The performance data presented (Sensitivity, Specificity, PPV, NPV, Correlation, Bland & Altman) directly reflects the performance of the "Compass automatic scoring" against manual PSG scoring. This is a measure of the algorithm's performance without a human in the loop during the scoring validation itself. While the intended use allows for physician manual correction, the validation study focused on the algorithm's intrinsic accuracy.
7. The Type of Ground Truth Used
- Expert Consensus / Expert Interpretation with a single expert. The ground truth was established by "manually scored PSG" by "the same RPSG technician." This falls under expert interpretation, specifically by a single expert rather than a consensus of multiple experts.
8. The Sample Size for the Training Set
- Not specified. The document does not provide any information about the size or characteristics of a training set. The study focuses solely on the validation data.
9. How the Ground Truth for the Training Set Was Established
- Not applicable/Not specified. Since no information is given about a training set, the method for establishing its ground truth is also not mentioned.
Ask a specific question about this device
Page 1 of 1