(240 days)
PocketCPR Intended Use: To assist users in the performance of effective CPR on a patient 8 years or older.
Effective cardiopulmonary resuscitation (CPR), when administered quickly after the on-set of cardiac arrest, is effective at saving lives and enabling complete neurological recovery, especially when followed by early defibrillation. The PocketCPR® was developed to provide users with real-time feedback during the administration of CPR, to assist CPR instructors during class exercises, and to provide CPR students with verbal instructions and feedback for at-home practice. When the PocketCPR is turned on, the default protocol being driven by the 2010 American Heart Association's (AHA) Guidelines, reminds the user to call for help. Once placed on the patient's chest and chest compressions have begun, PocketCPR generates a metronome at the recommended rate of chest compression, gives verbal and visual feedback on the quality of CPR being delivered and reminds the user to give two breaths after every 30 compressions. Measurement of actual chest wall movement is made by a state-of-the-art accelerometer micro-chip and microprocessor performing continuous analysis.
The provided text describes the PocketCPR® device, an aid for Cardiopulmonary Resuscitation (CPR), and its substantial equivalence to a predecessor device. However, it does not include specific acceptance criteria (e.g., target performance metrics like sensitivity, specificity, accuracy for CPR feedback) or a detailed study proving the device meets such criteria.
The bulk of the "Types of Testing" section lists various bench tests, simulated use on a manikin, usability studies, and biocompatibility, but it doesn't provide performance results against specific, quantifiable acceptance criteria.
Therefore, I cannot fully answer your request based on the provided text for certain points as the information is not present. I will indicate where information is missing.
Here's a breakdown of the available information:
1. Table of Acceptance Criteria and Reported Device Performance
Acceptance Criteria for CPR Feedback: Specific numerical acceptance criteria (e.g., what percentage accuracy is required for depth, rate, or release feedback) are not explicitly stated in the provided document. The document states the device "generates a metronome at the recommended rate of chest compression, gives verbal and visual feedback on the quality of CPR being delivered and reminds the user to give two breaths after every 30 compressions." This describes the functionality but not the performance targets or the measured performance against those targets.
Reported Device Performance: Similarly, specific quantitative performance data (e.g., measured accuracy of depth detection, rate measurements, or percentage of correct feedback instances) is not reported. The document notes "Measurement of actual chest wall movement is made by a state-of-the-art accelerometer micro-chip and microprocessor performing continuous analysis," but actual performance results are absent.
| Feature/Metric | Acceptance Criteria (As stated in document) | Reported Device Performance (As stated in document) |
|---|---|---|
| CPR Depth Feedback Accuracy | Not explicitly stated (Implied: "gives verbal and visual feedback on the quality of CPR being delivered" meeting AHA 2010 Guidelines) | Not reported quantitatively. States "Measurement of actual chest wall movement is made by a state-of-the-art accelerometer micro-chip and microprocessor performing continuous analysis." |
| CPR Rate Feedback Accuracy | Not explicitly stated (Implied: "generates a metronome at the recommended rate of chest compression" meeting AHA 2010 Guidelines) | Not reported quantitatively. |
| Full Release Detection Accuracy | Not explicitly stated (Implied: "allow for 'Full Release' on 'UP' stroke" feedback) | Not reported quantitatively. |
| Breaths Prompt Accuracy | Not explicitly stated (Implied: "reminds the user to give two breaths after every 30 compressions" meeting AHA 2010 Guidelines) | Device "maintains communicating 'Open Airway - Give Two Breaths' after each sequence of 30 compressions." |
| Orientation Guidance Accuracy | Not explicitly stated (Implied: "assist the caregiver in selecting proper orientation") | Device "will also prompt 'Analysis Halted' if improper orientation has occurred and will stay silent awaiting proper orientation." |
2. Sample size used for the test set and the data provenance
- Test Set Description: The document mentions "Simulated Use on Computerized Manakin for Data Collection" and "Usability Studies performing CPR using PocketCPR."
- Sample Size for Test Set: The sample size for these tests is not specified.
- Data Provenance: The document does not specify the country of origin for the data or whether the studies were retrospective or prospective. Given the mention of "Computerized Manakin" and "Usability Studies," it's highly likely the data was collected prospectively in a controlled environment.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
This information is not provided in the document. The type of ground truth used (manikin performance, expert observation) is not detailed, nor is the number or qualifications of any experts involved in assessing performance.
4. Adjudication method (e.g., 2+1, 3+1, none) for the test set
This information is not provided in the document.
5. If a multi-reader multi-case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance
- MRMC Study: An MRMC comparative effectiveness study, which typically compares human performance with and without AI, is not described in the document. The device is a direct feedback tool, not an "AI" in the typical sense of interpreting complex diagnostic images or data for a human reader. Its function is to directly guide the user.
- Effect Size: Therefore, effect size related to human reader improvement with AI assistance is not applicable or reported here.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- Standalone Performance: The document mentions "Measurement of actual chest wall movement is made by a state-of-the-art accelerometer micro-chip and microprocessor performing continuous analysis." While this implies an internal algorithm for measurement, no standalone performance evaluation of this algorithm's accuracy (e.g., against a gold standard for force or depth measurement) is explicitly detailed with specific metrics. The testing rather focuses on the device's output (feedback) in simulated use.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
- The document implies that the ground truth for the "Simulated Use on Computerized Manakin for Data Collection" and "Usability Studies" likely involves:
- Manikin-based measurements: Computerized manikins are designed to accurately measure CPR parameters (depth, rate, recoil). This would serve as a form of objective, engineered ground truth.
- Compliance with AHA Guidelines: The ultimate "ground truth" for the device's feedback is its adherence to the 2010 American Heart Association (AHA) Guidelines for CPR.
8. The sample size for the training set
- The document describes the device as being "programmed to meet the 2010 AHA Guidelines." It does not mention a "training set" in the context of machine learning. The device's "programming" is based on the AHA guidelines rather than learned from a dataset.
- Therefore, a "training set sample size" in the conventional machine learning sense is not applicable or provided.
9. How the ground truth for the training set was established
- As above, there is no mention of a "training set" for a machine learning model. The device's operational logic and feedback parameters are derived directly from the 2010 American Heart Association (AHA) Guidelines for CPR. These guidelines represent the accepted "ground truth" for effective CPR practices.
§ 870.5210 Cardiopulmonary resuscitation (CPR) aid.
(a)
CPR aid without feedback —(1)Identification. A CPR aid without feedback is a device that performs a simple function such as proper hand placement and/or simple prompting for rate and/or timing of compressions/breathing for the professionally trained rescuer, but offers no feedback related to the quality of the CPR being provided. These devices are intended for use by persons professionally trained in CPR to assure proper use and the delivery of optimal CPR to the victim.(2)
Classification. Class I (general controls). The device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 870.9.(b)
CPR aid with feedback —(1)Identification. A CPR Aid device with feedback is a device that provides real-time feedback to the rescuer regarding the quality of CPR being delivered to the victim, and provides either audio and/or visual information to encourage the rescuer to continue the consistent application of effective manual CPR in accordance with current accepted CPR guidelines (to include, but not be limited to, parameters such as compression rate, compression depth, ventilation, recoil, instruction for one or multiple rescuers, etc.). These devices may also perform a coaching function to aid rescuers in the sequence of steps necessary to perform effective CPR on a victim.(2)
Classification. Class II (special controls). The special controls for this device are:(i) Nonclinical performance testing under simulated physiological or use conditions must demonstrate the accuracy and reliability of the feedback to the user on specific compression rate, depth and/or respiration over the intended duration, and environment of use.
(ii) Labeling must include the clinical training, if needed, for the safe use of this device and information on the patient population for which the device has been demonstrated to be effective (including patient size and/or age limitations,
e.g., adult, pediatric and/or infant).(iii) For devices that incorporate electrical components, appropriate analysis and testing must demonstrate that the device is electrically safe and electromagnetically compatible in its intended use environment.
(iv) For devices containing software, software verification, validation, and hazard analysis must be performed.
(v) Components of the device that come into human contact must be demonstrated to be biocompatible.
(vi) Human factors testing and analysis must validate that the device design and labeling are sufficient for effective use by the intended user.
(3)
Premarket notification. The CPR Aid with feedback device is exempt from the premarket notification procedures in subpart E of part 807 of this chapter if it does not contain software (e.g., is mechanical or electro-mechanical) and is in compliance with the special controls under paragraph (b)(2) of this section, subject to the limitations of exemptions in § 870.9.