(74 days)
The ORG-9700A Multiple Patient Receiver is intended for use by medical professionals with Nihon Kohden telemetry transmitters and central stations to provide cardiac and vital signs monitoring for multiple patients within a medical facility. The device detects patient vital sign alarm conditions and includes an algorithm to detect cardiac arrhythmias. The intended use of the modified device has not changed as a result of the modifications. The device is available for use on all patient populations.
New ORG-9700A has the same intended use as the previously marketed telemetry system, which is for use by medical professionals with Nihon Kohden telemetry transmitters and central stations to provide cardiac and vital signs monitoring for multiple patients. The device can receive the patient's ECG, respiration, SpO2, and non-invasive blood pressure data (NIBP value is only displayed on the central monitor) from a transmitter and send it to Nihon Kohden Central monitor within the medical facility. The device is designed as a component of a central monitor network to be used in the ICU, CCU, HCU, recovery room, operating room and general ward. The device will receive and transmit physiological data from telemetry transmitters/receivers and generate an alarm when a measured parameter falls outside a preset limit or when an arrhythmia is detected.
The device is not sterile. The device is not directly connected to patients. It is used as a central monitoring system for obtaining a series of patient vital information.
The provided documentation for the NIHON KOHDEN AMERICA, INC. ORG-9700A Multiple Patient Receiver (K071058) does not contain detailed information regarding acceptance criteria or a specific study proving the device meets these criteria in the way typically required for AI or algorithmic performance evaluation.
The submission is a 510(k) for a telemetry receiver, which is a hardware device for signal exchange and monitoring, not an AI or diagnostic algorithm. The type of acceptance criteria and study design you're asking for (e.g., sample size for test sets, data provenance, expert adjudication, MRMC studies, standalone performance of an algorithm, ground truth methods for training sets) are specific to evaluating the clinical performance of diagnostic algorithms or decision support systems, especially those using AI.
For this device, the "performance" described pertains to its functionality as a signal receiver and transmitter within a medical facility, and its ability to detect patient vital sign alarm conditions and include an algorithm to detect cardiac arrhythmias.
Here's an analysis based on the information provided, highlighting why many of your requested details are not present in this type of submission:
1. A table of acceptance criteria and the reported device performance
The document states:
- "The device was subject to electromagnetic, environmental, safety and performance testing procedures."
- "Software validation was satisfactory completed as part of product's design validation."
- "The results confirmed that the device performed within specifications."
Acceptance Criteria & Device Performance (as inferred from the document):
| Acceptance Criteria (Inferred) | Reported Device Performance (Inferred) |
|---|---|
| Compliance with Electromagnetic Standards | Performed within specifications (implied) |
| Compliance with Environmental Standards | Performed within specifications (implied) |
| Compliance with Safety Standards | Performed within specifications (implied) |
| Performance within Specifications (General) | Performed within specifications (explicitly stated) |
| Software Validation Satisfactory | Satisfactorily completed (explicitly stated) |
| Accurate Detection of Patient Vital Sign Alarm Conditions | Device detects alarm conditions (stated intent) |
| Functionality of Cardiac Arrhythmia Detection Algorithm | Device includes such an algorithm (stated intent) |
| Substantial Equivalence to Predicate Device | Determined to be substantially equivalent by FDA |
Note: The actual specifications and detailed performance metrics (e.g., signal-to-noise ratio, alarm accuracy rates, arrhythmia detection sensitivity/specificity) are not provided in this summary document. These would typically be found in the more detailed sections of the 510(k) submission, not in the public summary.
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
This information is not provided in the given 510(k) summary. For a hardware device performing signal transmission and basic alarm/arrhythmia detection, performance testing would typically involve engineering verification and validation, rather than clinical studies with large patient test sets in the context of diagnostic accuracy. The testing would focus on the hardware's reliability, signal integrity, and the correct functioning of the embedded software/firmware for its stated purposes.
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 information is not provided and is unlikely to be relevant for the type of device being submitted. Ground truth (e.g., expert consensus on diagnoses) is crucial for evaluating human-in-the-loop diagnostic studies or standalone AI diagnostic algorithms. For a telemetry receiver, ground truth would relate to the accuracy of physiological signal representation and the correct triggering of alarms based on predefined thresholds, which are verified through technical testing, not necessarily by expert clinical review of specific patient cases in a study.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
This information is not provided and is not applicable to a hardware device submission of this nature. Adjudication methods are used in clinical studies to resolve discrepancies among expert readers when establishing ground truth for diagnostic accuracy.
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
An MRMC study was not conducted (or at least not reported in this summary). This type of study is relevant for evaluating the impact of an AI-powered diagnostic tool on human reader performance, which is not the primary function or claim of this telemetry receiver. The device "includes an algorithm to detect cardiac arrhythmias," but its primary role is signal reception and transmission, with this algorithm as a feature, not necessarily a standalone diagnostic AI system being evaluated for comparative effectiveness with human readers.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
While the device includes an "algorithm to detect cardiac arrhythmias," the summary does not report specific standalone performance metrics for this algorithm (e.g., sensitivity, specificity, accuracy against a gold standard dataset of arrhythmia events). The statement "Software validation was satisfactory completed as part of product's design validation" implies that the algorithm's performance was tested as part of the overall system, but no details of that testing are provided in this summary.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)
This information is not provided and for this device, a "ground truth" in the clinical sense (e.g., pathology for a cancer diagnosis) is less applicable. For the arrhythmia detection algorithm, the ground truth would likely be established through:
- Simulated physiological signals: Testing the algorithm with known-arrhythmic and non-arrhythmic signal patterns.
- Reference ECG recordings: Comparing the algorithm's output against interpretations by expert cardiologists or established annotated databases.
- However, specific details are not mentioned in the summary.
8. The sample size for the training set
This information is not provided. As this is primarily a hardware device with an embedded algorithm (likely rule-based or a simpler classifier, rather than a deep learning model requiring a large training set), the concept of a "training set" might not be applicable in the same way it is for modern AI. If it were a sophisticated AI algorithm, details about the training data would be crucial.
9. How the ground truth for the training set was established
This information is not provided, as the details about a training set are absent.
In summary:
The provided 510(k) summary focuses on demonstrating substantial equivalence to a predicate device for a telemetry receiver. The performance data discussed relates to engineering and software validation (electromagnetic, environmental, safety, general performance specifications). It does not provide the kind of detailed clinical study data, ground truth establishment methods, or AI-specific performance metrics that would be expected for a diagnostic AI algorithm or a device requiring extensive clinical validation for diagnostic accuracy claims. The "algorithm to detect cardiac arrhythmias" is mentioned as a feature, but its specific performance evaluation details are not part of this high-level summary.
§ 870.2910 Radiofrequency physiological signal transmitter and receiver.
(a)
Identification. A radiofrequency physiological signal transmitter and receiver is a device used to condition a physiological signal so that it can be transmitted via radiofrequency from one location to another, e.g., a central monitoring station. The received signal is reconditioned by the device into its original format so that it can be displayed.(b)
Classification. Class II (performance standards).