(77 days)
The device is intended to be used by medical professionals only for the measurement of systolic and diastolic blood pressure values. The device is intended to be used with blood pressure cuffs supplied for use with the device which are manually attached to a patient and manually inflated, with a manual method for detecting Korotkoff sounds. The device cuffs are available in a range of adult, pediatric and neonate sizes for use in exactly the same way and substantially equivalent to the Accoson greenlight 300 sphygmomanometer under 510(k) number K970111 for adult, pediatric and neonate patient groups.
Accoson Greenlight 300 Sphygmomanometer
This is challenging because the provided text is a 510(k) clearance letter for a sphygmomanometer, which is a very different type of device than what is typically associated with AI/ML and the requested information (e.g., acceptance criteria, study details, ground truth, training sets).
The document is for an Accoson Greenlight 300 Sphygmomanometer, which is a non-invasive blood pressure measurement system. This is a purely mechanical/electronic device, not an AI/ML-driven one. Therefore, the concepts of "AI assistance," "ground truth establishment" for training, "MRMC studies," or "standalone algorithm performance" are not applicable to this device.
The 510(k) clearance process for such a device primarily focuses on demonstrating "substantial equivalence" to a legally marketed predicate device, often through performance testing that adheres to recognized standards (e.g., AAMI, ISO for blood pressure monitors). The letter itself does not contain the detailed study results or acceptance criteria, but rather references the submission made by the manufacturer.
Given this, I cannot directly extract the requested information in the format provided because the device and context are inappropriate for those questions. Instead, I can explain what typically would be expected for a device like this based on common regulatory practices for sphygmomanometers, but this would be inferred and not directly from the provided text.
Based on the typical regulatory requirements for a sphygmomanometer (and what would be found in a 510(k) submission, though not in the letter provided):
1. A table of acceptance criteria and the reported device performance
-
Acceptance Criteria (Example for a traditional sphygmomanometer, often based on AAMI/ISO standards):
- Accuracy (Mean Difference): ≤ ±5 mmHg
- Accuracy (Standard Deviation): ≤ 8 mmHg
- Device Measurement Range: Typically 0-300 mmHg (systolic and diastolic)
- Cuff Sizing: Adherence to adult, pediatric, and neonate ranges (as stated in the Indications for Use).
- Inflation/Deflation Rates: Within specified tolerances.
- Durability/Environmental Testing: Device maintains accuracy and function after specified tests (e.g., shock, vibration, temperature, humidity).
- Safety Standards: Compliance with electrical safety (e.g., IEC 60601-1) and biocompatibility for materials in contact with skin.
- Reproducibility: Within specified limits across multiple measurements.
-
Reported Device Performance: This information would be in the detailed testing reports submitted by A. C. Cossor & Son (Surgical) Ltd. to the FDA, not in the clearance letter. The letter only states that the FDA "reviewed your Section 510(k) premarket notification... (and) found the device is substantially equivalent." This implies the performance met the specified criteria.
2. Sample size used for the test set and the data provenance
- Sample Size (Clinical Accuracy Testing): For clinical validation of blood pressure devices (e.g., according to ISO 81060-2 or similar standards), typically 85 subjects are required for validation against a reference method (e.g., auscultation by trained observers). This would include a diverse range of blood pressures.
- Data Provenance: Not specified in the letter. Likely collected prospectively during a clinical study conducted by the manufacturer, possibly in the UK or another country.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts
- Number of Experts: For clinical validation using auscultation (the "gold standard" for manual BP), typically two or three trained observers (a.k.a. "experts") are used simultaneously to listen to Korotkoff sounds.
- Qualifications: These would be healthcare professionals (e.g., physicians, nurses, trained technicians) with demonstrated proficiency in indirect blood pressure measurement using a mercury sphygmomanometer and stethoscope, adhering to standardized protocols. They would be trained and show inter-rater reliability.
4. Adjudication method for the test set
- Adjudication Method (Clinical BP Validation): When using multiple observers for auscultation, it's common for the average of their readings (if within a specified tolerance, e.g., 2-4 mmHg) to serve as the reference standard (ground truth). If readings differ significantly, either a third observer is used, or the measurement is discarded and repeated. This isn't strictly an "adjudication" in the AI sense, but rather a method for establishing a reliable reference measurement.
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 a traditional sphygmomanometer. There is no AI component, and thus no "human readers improving with AI assistance."
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- Not Applicable. This device is a traditional sphygmomanometer. Its "standalone" performance is its fundamental operation. There is no separate algorithm to be evaluated.
7. The type of ground truth used
- Ground Truth: For accuracy validation, the ground truth would be simultaneous direct arterial blood pressure measurements (invasive) or, more commonly for non-invasive devices, auscultatory measurements using a standardized mercury sphygmomanometer by trained observers (consensus).
8. The sample size for the training set
- Not Applicable. This device does not use a "training set" in the context of machine learning or AI.
9. How the ground truth for the training set was established
- Not Applicable. No training set for AI/ML.
In summary, the provided document is a 510(k) clearance for a basic medical device (a sphygmomanometer), not an AI/ML product. Therefore, most of the questions related to AI/ML study methodologies are not relevant and cannot be answered from this specific document.
§ 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).