(121 days)
The Circadia C100 System is indicated for both contactless spot checking and continuous measurement of respiratory rate data as part of a vital signs assessment. The system records, transmits, and displays respiratory rate from multiple connected devices for retrospective analysis only. The system is intended to be used under the care of clinicians and medically qualified personnel.
The system is indicated for use in adult patients during no-motion conditions, for patients in health care facilities. It is available for sale only upon the order of a physician or licensed health care provider.
The Circadia C100 System is not indicated for active patient monitoring, as it does not provide alarms for timely response in life-threatening situations. It is not intended to monitor vital signs. This system is not an apnea monitor.
The Circadia C100 System consists of three components:
- "The Monitor" - Circadia Contactless Breathing Monitor: The Monitor is a contact-less breathing monitor placed at a bedside table. The Monitor records respiratory rate of adult patients in hospitals and clinical settings (e.g. post-acute care settings such as General Care Floors (GCF), Inpatient Rehabilitation Facilities (IRF), and Skilled Nursing Facilities (SNF)). The Monitor utilizes a UWB-based (Ultra-Wideband based) motion sensor to detect the movement of the abdomen and the chest to derive respiratory data.
- "The App" - Circadia Pro Android App: The App allows Technical Support Staff to setup and configure monitors and allows Clinical Staff to review data collected by the monitors. The App is operated from the central "hub" of a ward or nursing station to monitor data collected by multiple monitors for retrospective analysis only, not for active patient monitoring. The App operates from an Android tablet and uses a USB-C connection for initial configuration. The App is designed to communicate with The Monitor and allow it to connect to a Wi-Fi network and subsequently, The Cloud Service.
- "The Cloud Service" - Circadia Cloud Service: The Cloud Service offers a set of APIs that allows the monitor to connect to server, send respiratory data over a secure channel, and allows the monitor to track Clinical Staff's personal data and the patient's respiratory data.
Here's a breakdown of the acceptance criteria and study that proves the device meets them, based on the provided FDA 510(k) summary for the Circadia C100 System:
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria for the Circadia C100 System, specifically for its respiratory rate (RR) accuracy, are derived from the clinical testing results that demonstrated its substantial equivalence to a predicate device.
| Acceptance Criterion (from testing demonstrating substantial equivalence) | Reported Device Performance |
|---|---|
| Accuracy of Respiration rate: +/- 2 breaths/min | Achieved 88.2% accuracy |
| 95% limits of agreement: Within an acceptable range relative to predicate/gold standard | -3.2 to +0.7 breaths/min |
Note: The document states that the performance met 'acceptance' but does not explicitly list the target acceptance criteria in the format of a table with specific thresholds. Instead, it presents the results of a comparison study intended to demonstrate substantial equivalence to a predicate device and a gold standard. The +/- 2 breaths/min accuracy and the -3.2 to +0.7 breaths/min 95% Limits of Agreements (LoA) are the results that satisfied the FDA's requirements for substantial equivalence for this device.
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size:
- Comparison to end-tidal CO2 capnography (gold standard): 12 patients
- Comparison to chest and abdomen worn band (predicate/reference device): 26 patients
- Total unique patients: While not explicitly stated, it seems these are two distinct sub-studies, leading to a total of 38 unique subjects (12 + 26).
- Data Provenance: Not specified (e.g., country of origin, specific healthcare facilities). The document mentions "patients in health care facilities" (page 2) and "adult patients in hospitals and clinical settings" (page 4), but no specific location.
- Retrospective or Prospective: Not explicitly stated. However, clinical testing for 510(k) submissions typically involves prospective data collection to evaluate real-time performance. Given the details about monitoring various conditions and subject postures, it suggests prospective data collection.
3. Number of Experts Used to Establish Ground Truth and Qualifications
- Number of Experts: Not specified.
- Qualifications of Experts: Not specified. The ground truth for the capnography data is implicitly established by the device producing that data ("manually scored end-tidal CO2 capnography data"). For the chest/abdomen worn band, it serves as a comparative reference rather than a direct ground truth source that required expert interpretation for its output.
4. Adjudication Method for the Test Set
- Not applicable/Not specified. The ground truth was established by readings from established medical devices (end-tidal CO2 capnography and a chest/abdomen worn band), not through expert consensus requiring adjudication.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
- No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done.
- The study design focused on comparing the device's measurements against established methods (gold standard capnography and a predicate band), rather than evaluating how AI (or the device's) assistance improves human reader performance. This device is a measurement device, not an AI-assisted diagnostic tool for human readers.
6. Standalone (Algorithm Only) Performance
- Yes, a standalone performance evaluation was done. The "accuracy of Respiration rate" and "95% limits of agreements" reported are for the Circadia C100 system itself, directly measuring the respiratory rate. The clinical testing specifically describes comparing the device's performance against the gold standard and the predicate device.
7. Type of Ground Truth Used
- Clinical/Physiological Reference Data: The primary ground truth was "manually scored end-tidal CO2 capnography data" (from devices cleared in K040875 and K120888), which is considered a gold standard for respiratory rate measurement.
- Comparative Reference: Performance was also compared against data from a "chest and abdomen worn band" (device cleared in K122516), which serves as a predicate/reference device rather than a ground truth in itself.
8. Sample Size for the Training Set
- Not specified. This document is a 510(k) summary for a medical device that uses UWB-based motion sensing, not a machine learning or AI algorithm in the contemporary sense that typically relies on large training datasets for feature learning. While the device certainly processes signals, the document doesn't indicate a machine learning training phase with a distinct training set. The device likely uses signal processing algorithms tuned during development rather than trained on a large dataset in the way a deep learning model would be.
9. How the Ground Truth for the Training Set Was Established
- Not applicable, as a distinct training set with established ground truth as for a machine learning model is not mentioned or implied by the description of the device's technology. The device utilizes physical principles (UWB radar to detect chest/abdomen movement) and signal processing, not a data-driven "training" process in the AI/ML context.
§ 868.2375 Breathing frequency monitor.
(a)
Identification. A breathing (ventilatory) frequency monitor is a device intended to measure or monitor a patient's respiratory rate. The device may provide an audible or visible alarm when the respiratory rate, averaged over time, is outside operator settable alarm limits. This device does not include the apnea monitor classified in § 868.2377.(b)
Classification. Class II (performance standards).