(448 days)
The Connex Central Station is intended to be used by clinicians for the central monitoring of neonatal, pediatric and adult patients in health care facilities. In addition to the central monitoring of patient data and alarms, the Connex software can include optional modules to provide extended recording of patient data, including full disclosure.
Both the Welch Allyn primary predicate and subject Connex Central Station devices are software devices for a Windows-based operating system that provide clinicians with a means to remotely monitor the health of several patients simultaneously.
This document is a 510(k) summary for the Welch Allyn Connex Central Station (v.1.8.5). It primarily focuses on demonstrating substantial equivalence to a predicate device, not on presenting a detailed study proving performance against specific acceptance criteria for a new AI/ML device.
Therefore, many of the requested details regarding acceptance criteria, study design, expert qualifications, and ground truth establishment for an AI/ML model are not present in this document. The document states "No clinical studies were utilized for the purpose of obtaining safety and effectiveness data." and refers to software verification and validation, and adherence to various medical device standards.
However, based on the provided text, I can infer and extract some relevant information as best as possible, particularly regarding the non-clinical testing performed and the nature of the device.
Here's an attempt to answer your questions based on the provided text, noting where information is explicitly not available for an AI/ML context:
Acceptance Criteria and Device Performance (Inferred)
Since this is a submission for substantial equivalence based on a predicate device and not a new AI/ML algorithm requiring performance metrics like sensitivity/specificity against a ground truth, the "acceptance criteria" here are primarily met through software verification and validation testing and adherence to recognized medical device standards. The performance is not reported as specific clinical metrics but rather as meeting the functional and safety requirements for a central patient monitoring station.
The key change is the addition of ECG parameter display from another cleared device (Welch Allyn CVSM, K171621), and the acceptance is that this integration does not introduce new safety or effectiveness concerns.
| Acceptance Criteria Category (Inferred from standards) | Reported Device Performance (Summary from document) |
|---|---|
| Software Functionality & Performance: | Met design requirements and performance, functionality characteristics. The methods for displaying ECG parameters are the same as previously cleared parameters. |
| Usability Engineering (IEC 62366-1): | Tested (implies acceptance criteria met). |
| Basic Safety & Essential Performance (IEC 60601-1-8): | Tested (implies acceptance criteria met). |
| Software Life Cycle Processes (IEC 60304): | Tested (implies acceptance criteria met); classified as "Major" level of concern. |
| Electrocardiographic Monitoring Equipment Safety (IEC 60601-2-27): | Tested (implies acceptance criteria met). |
| Risk Management (ISO 14971, AAMI 80001-1): | Tested (implies acceptance criteria met). |
| Labeling and Information (ISO 15223-1): | Tested (implies acceptance criteria met). |
| Substantial Equivalence: | Concluded that the device is substantially equivalent to the primary predicate, with the added ECG feature not altering safety/effectiveness. |
2. Sample size used for the test set and the data provenance:
- Test Set: Not applicable in the context of an AI/ML test set with patient data. The "testing" here refers to software verification and validation activities and compliance with standards. There's no mention of a "test set" of patient data for performance evaluation in the way an AI/ML algorithm would use it.
- Data Provenance: Not applicable. The document refers to "well-established, scientific methods" for evaluating new ECG parameter features, but this is about the display of parameters already collected by another cleared device (CVSM, K171621), not new data analysis or inference from a patient data set by the Connex Central Station itself.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Not applicable. This submission does not involve establishing ground truth for an AI/ML algorithm's clinical performance. The ground truth for the ECG parameters themselves would have been established during the clearance of the Welch Allyn CVSM (K171621), but that's not detailed here.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:
- Not applicable. This is not an AI/ML study involving human readers and adjudicated ground truth.
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:
- No. The document explicitly states: "No clinical studies were utilized for the purpose of obtaining safety and effectiveness data."
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
- Not applicable. This device integrates and displays data from other cleared devices; it's a central monitoring station, not an AI/ML algorithm that operates "standalone" to produce a diagnostic or prognostic output from raw data. Its "performance" is based on its ability to accurately receive, process, and display parameters from connected devices and manage alarms, complying with relevant standards.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- Not applicable in the context of an AI/ML performance study for this submission. The "ground truth" for the displayed ECG parameters originates from the cleared Welch Allyn CVSM device (K171621), which would have undergone its own validation. For the central station itself, the "ground truth" lies in its adherence to functional specifications, risk management, and the standards listed (e.g., IEC 62304 for software, IEC 60601-2-27 for ECG monitoring equipment safety aspects relevant to its display).
8. The sample size for the training set:
- Not applicable. This device is not an AI/ML algorithm that undergoes a training phase with a dataset.
9. How the ground truth for the training set was established:
- Not applicable. (See point 8).
Summary of what the document DOES tell us about the "study":
The "study" or evaluation performed was primarily:
- Software Verification and Validation (V&V): This involved testing the software against its design requirements and functional specifications. The "level of concern" for the software was "Major" (meaning a failure could result in serious injury or death).
- Compliance with Recognized Standards: The device was tested to evaluate its performance based on various international standards for medical devices, specifically:
- IEC 62366-1:2015 (Usability Engineering)
- IEC 60601-1-8:2012 (Basic Safety and Essential Performance)
- IEC 62304:2015 (Medical Device Software - Software Life Cycle Processes)
- IEC 60601-2-27:2011 (Electrocardiographic Monitoring Equipment Safety)
- ISO 14971:2019 (Risk Management)
- AAMI 80001-1:2010 (Risk Management for IT networks with medical devices)
- ISO 15223-1:2016 (Symbols for labeling)
- FDA Special Controls Guidance for Arrhythmia Detector and Alarm (October 28, 2003).
The key finding from this "study" or evaluation was that even with the new feature (displaying ECG parameters), the device continues to meet these standards and its design requirements, and therefore remains substantially equivalent to its predicate, raising no new questions of safety or effectiveness.
§ 870.2300 Cardiac monitor (including cardiotachometer and rate alarm).
(a)
Identification. A cardiac monitor (including cardiotachometer and rate alarm) is a device used to measure the heart rate from an analog signal produced by an electrocardiograph, vectorcardiograph, or blood pressure monitor. This device may sound an alarm when the heart rate falls outside preset upper and lower limits.(b)
Classification. Class II (performance standards).