(269 days)
The Electronic Blood Pressure Monitor is intended to measure the systolic and diastolic blood pressure as well as the pulse rate of adult person via non-invasive oscillometric technique in which an inflatable cuff is wrapped around the upper arm. It can be used at medical facilities or at home. The intended arm circumference includes 22 cm32 cm and 32 cm42 cm.
The proposed device, Electronic Blood Pressure Monitor, is a battery driven automatic non-invasive blood pressure monitor. It can automatically complete the inflation and measurement, which can measure systolic and diastolic blood pressure and pulse rate of the adult person at upper arm within its claimed range and accuracy via the oscillometric technique. User can select the unit of the measurement: mmHg or KPa.
The device has the data storage function in order for data reviewing, including the systolic pressure, diastolic pressure, pulse rate and measurement time. It has a bar indicating function, which can indicate the WHO (World Health Organization) Blood Pressure Classification of the measured blood pressure by referencing Diastolic Blood Pressure issued at Journal of Hypertension 1999. Vol 17, No.2.
The proposed electronic blood pressure monitor has 12 models, including PG-800B22, PG-800B23, PG-800B26, PG-800B27, PG-800B31, PG-800B32, PG-800B33, PG-800B35, PG-800B36, PG-800B37, PG-800B42 and PG-800B43. All models follow the same software, measurement principle and NIBP algorithm. The main differences are product appearance and key numbers.
The proposed device is intended to be used in medical facilities or at home.
The product is provided non-sterile, and not to be sterilized by the user prior to use.
The provided document is a 510(k) summary for the "Electronic Blood Pressure Monitor" by Shenzhen Pango Electronic Co., Ltd. It declares substantial equivalence to a predicate device. However, this document primarily focuses on regulatory compliance, outlining intended use, device description, comparison to a predicate, and non-clinical test conclusions based on various IEC standards.
Crucially, this document does NOT contain information about a clinical study involving a test set, expert readers, ground truth establishment, or any data related to diagnostic performance metrics (e.g., sensitivity, specificity, accuracy) that would be relevant for a typical AI/ML medical device submission.
Instead, the performance of this blood pressure monitor is assessed against standards for accuracy defined by organizations like ISO and IEC for non-invasive sphygmomanometers. Therefore, the "acceptance criteria" and "proof of meeting criteria" here refer to the device's ability to measure blood pressure within the specified accuracy limits of these standards, not against a human expert consensus.
Given this, I will interpret the requested information in the context of a blood pressure monitor's accuracy testing, as implied by the reference to IEC 80601-2-30:2009 and ISO 81060-2:2013 (which generally address accuracy requirements for automated sphygmomanometers).
Based on the provided document, here's an attempt to answer your questions, with the understanding that the "study" is likely a performance verification against a standard, not a clinical trial with human readers assisting AI.
1. A table of acceptance criteria and the reported device performance
The document states that the device "Comply with IEC 80601-2-30:2009 and ISO 81060-2:2013" for "Particular Performance." These standards define the accuracy requirements for automated non-invasive sphygmomanometers.
Acceptance Criteria (as per ISO 81060-2:2013 for clinical validation for blood pressure devices):
The ISO 81060-2:2013 standard typically requires two main criteria for accuracy:
- Mean difference: The mean difference between the device's measurement and the reference standard (e.g., mercury sphygmomanometer) should be $\le \pm 5$ mmHg.
- Standard deviation: The standard deviation of the difference between the device's measurement and the reference standard should be $\le 8$ mmHg.
- Individual differences (AAMI/BHS protocols often append this): A high percentage of measurements (e.g., >60% for AAMI, or various grades for BHS) must fall within certain error margins (e.g., $\pm 5, \pm 10, \pm 15$ mmHg).
Reported Device Performance:
The document states: "The test results demonstrated that the proposed device complies with the following standards: ... IEC 80601-2-30:2009, Medical electrical equipment - Part 2-30: Particular requirements for the basic safety and essential performance of automated noninvasive sphygmomanometers" and "Comply with IEC 80601-2-30:2009 and ISO 81060-2:2013".
Therefore, the reported device performance is that it met the accuracy requirements outlined in these standards. The specific numerical values (mean difference and standard deviation) are not provided in this 510(k) summary, but the declaration of compliance serves as the proof.
| Acceptance Criteria (from ISO 81060-2:2013) | Reported Device Performance |
|---|---|
| Mean difference $\le \pm 5$ mmHg | Complies with ISO 81060-2:2013 (indicating these criteria were met) |
| Standard deviation $\le 8$ mmHg | Complies with ISO 81060-2:2013 (indicating these criteria were met) |
| (Additional criteria for individual differences vary by protocol, but compliance with the standard implies these were also met) | Complies with ISO 81060-2:2013 (indicating these criteria were met) |
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
The document does not specify the sample size for the accuracy testing. ISO 81060-2:2013 typically recommends a minimum of 85 participants for a clinical validation study to assess accuracy. The data provenance (country of origin, retrospective/prospective) is also not disclosed in this summary. Such details would usually be found in the full test report, not the summary.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)
This is not applicable in the context of a blood pressure monitor's accuracy testing. The "ground truth" for blood pressure measurements is typically established by simultaneous measurements using a reference standard device, often a double-headed mercury sphygmomanometer, read by trained observers, not "experts" in the sense of medical specialists adjudicating an image.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
Not applicable. Blood pressure accuracy testing does not involve adjudication of readings in the typical sense of expert consensus on an image or medical condition. Instead, it involves comparison to standard reference measurements.
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
Not applicable. This device is an automatic blood pressure monitor, not an AI-assisted diagnostic tool that human readers would use. There is no human-in-the-loop component or AI assistance for interpretation.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
Yes, the fundamental performance assessment of an automated blood pressure monitor is a standalone performance evaluation. The device (algorithm and hardware) measures blood pressure and provides a reading without human intervention in the measurement process. The "standalone" performance here refers to its accuracy against a reference standard. The document confirms that "Non clinical tests were conducted to verify that the proposed device met all design specifications as was Substantially Equivalent (SE) to the predicate device. The test results demonstrated that the proposed device complies with the following standards."
7. The type of ground truth used (expert concensus, pathology, outcomes data, etc)
For blood pressure monitor accuracy, the ground truth is established by simultaneous, direct comparison to a validated reference measurement method, typically a mercury sphygmomanometer or an equivalent standard, performed by two trained observers. It is a live physiological measurement, not an static image or pathology result.
8. The sample size for the training set
This device does not appear to be an AI/ML model that undergoes "training" in the conventional sense of deep learning. It uses an "oscillometric technique" and an "NIBP algorithm" (Non-Invasive Blood Pressure algorithm). While such algorithms might be refined and calibrated, the concept of a "training set" as understood in modern AI/ML development (with vast amounts of labeled data) is not explicitly stated or implied by the document for this device. It's more of a calibration and validation process against physical principles and existing data.
9. How the ground truth for the training set was established
As there's no explicitly mentioned "training set" in the AI/ML sense, this question isn't directly applicable. The underlying algorithm would have been developed based on general principles of oscillometry and vast amounts of existing physiological data, where the "ground truth" would originate from reference blood pressure measurements across diverse populations.
§ 870.1130 Noninvasive blood pressure measurement system.
(a)
Identification. A noninvasive blood pressure measurement system is a device that provides a signal from which systolic, diastolic, mean, or any combination of the three pressures can be derived through the use of tranducers placed on the surface of the body.(b)
Classification. Class II (performance standards).