Search Results

The Stasis Monitoring System is intended for use by clinicians and medically qualified personnel for sinqle or multi-parameter vital signs monitoring of adult patients (≥21 years of age). It is indicated for 3 lead ECG, respiration rate (RESP), heart rate (HR), noninvasive blood pressure (NIBP), noninvasive monitoring of functional oxygen saturation of arterial hemoglobin (SpOz), pulse rate (PR), and skin temperature (TEMP) in hospital-based facilities; including, general medical-surgical floors, intermediate care floors, and emergency departments. The Stasis Monitoring System includes bedside patient monitors that communicate with mobile tablets through wireless Bluetooth Low Energy (BLE) communication. The Stasis Monitoring System can generate alerts when rate-based cardiac arrhythmias such as asystole are detected, and when physiological vital signs fall outside of selected parameters.

The Stasis Monitoring System has a notification system that communicates data and alarms to a Stasis Tablet. It is intended to supplement the primary alarms which originate at the Stasis Monitor device.

The Stasis Monitoring System consists of a compact six-parameter vital signs monitor that sits at the patient's bedside and communicates via Bluetooth with off the shelf Android tablets running a custom Stasis application. The parameters measured include the following: 3 lead ECG, respiration rate (RESP), heart rate (HR), noninvasive blood pressure (NIBP), noninvasive monitoring of functional oxygen saturation of arterial hemoglobin (SpO2), pulse rate (PR), and skin temperature (TEMP). The System uses traditional wired sensor technology (i.e., sensors cables are connected to back of Monitor with sensors applied to patient) to acquire the vital signs. All sensors and cables used are off-the-shelf and are already 510(k) cleared as identified in the bulleted list on the following page below. The primary data display and control for the monitoring system is on the Android tablet (see Fig 1 below).

The provided document is a 510(k) summary for the Stasis Monitoring System. It describes the device, its intended use, and compares it to a predicate device to establish substantial equivalence.

Based on the nature of the device (a vital signs monitor) and the information provided, it's important to note that the acceptance criteria and study details requested are primarily focused on diagnostic or AI/machine learning devices that generate specific outputs requiring human-expert adjudication or comparison to a ground truth.

For a vital signs monitoring system like the Stasis Monitoring System, the "acceptance criteria" are typically related to the accuracy and precision of its measurements (e.g., heart rate, respiration rate, blood pressure, SpO2, temperature) compared to established reference methods or standards, and its ability to correctly generate alarms when vital signs fall outside set parameters. The "study" for such devices often involves bench testing with simulators and/or clinical performance studies to demonstrate measurement accuracy and alarm functionality.

The document refers to various tests performed, but does not provide granular details in the format requested for a diagnostic AI/ML device.

Here's an attempt to extract and interpret the information based on the provided text, addressing each point of your request as much as possible:

1. Table of acceptance criteria and the reported device performance

The document primarily focuses on demonstrating substantial equivalence to the predicate device, Sotera Wireless ViSi Mobile Monitoring System. This means the Stasis Monitoring System is expected to perform "as well as" or similarly to the predicate device. Specific numerical acceptance criteria are not explicitly stated for all parameters in a pass/fail outcome, but rather a comparison of specifications with the predicate.

Note on "Acceptance Criteria": For vital signs monitors, acceptance criteria for accuracy are often specified in relevant IEC standards (e.g., IEC 80601-2-56 for clinical thermometers, IEC 80601-2-27 for ECG, IEC 80601-2-30 for NIBP, etc.). The document states that these standards were used for testing (e.g., IEC 60601-2-27, IEC 80601-2-30, IEC 80601-2-56). The table above primarily lists the performance specifications rather than explicit "acceptance criteria" thresholds.

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 states:

• "Testing was conducted in-house by trained personnel using simulators to obtain the functional and accuracy test results."
• "Testing was also conducted by 3rd party trained personnel using simulators to obtain the functional and accuracy test results."

This indicates that the "test set" primarily consisted of simulated data generated by inanimate simulators, not human patient data.

• Sample size: Not specified in terms of number of simulated cases or data points.
• Data provenance: Not applicable in the context of human patients, as simulators were used. The testing was reported as "in-house" and by "3rd party."
• Retrospective or prospective: Not applicable, as simulated data was used.

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 question is not directly applicable in the typical sense for this device.

• The "ground truth" for vital signs measurements from simulators is the known output of the simulator itself, which is designed to produce precise and accurate physiological signals according to specifications.
• The document does not mention any human experts establishing ground truth for the simulator outputs; rather, the accuracy of the device's measurements would be compared against the simulator's known parameters.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

Not applicable. There was no mention of human adjudication for the simulator-based testing. The device's measurements were compared against the simulator's known values.

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. The Stasis Monitoring System is a vital signs monitor, not an AI-assisted diagnostic tool that interprets medical images or complex data for human readers. It provides raw vital sign data and alerts based on pre-set parameters. Therefore, an MRMC study comparing human performance with and without AI assistance is not relevant to this device.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

Yes, in a sense, the primary testing described is "standalone" performance of the device's measurement capabilities. The system was tested to measure vital signs and generate alerts directly from simulator inputs. The "human-in-the-loop" aspect here refers to clinicians using the monitor, but the accuracy tests evaluate the device's ability to measure and alert autonomously based on its programming and sensor inputs.

• "Testing for Stasis Monitoring System was performed to ensure that all functional requirements have been met, and that core functions execute as expected."
• "Testing was conducted in-house... using simulators to obtain the functional and accuracy test results."

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc)

The ground truth for the functional and accuracy testing was derived from simulators. The simulators provided known and controlled physiological signals or values against which the Stasis Monitoring System's measurements were compared.

8. The sample size for the training set

Not applicable. The Stasis Monitoring System is a traditional vital signs monitor. It does not appear to employ machine learning algorithms or AI that require a "training set" in the common computational sense (i.e., for learning patterns from data). Its logic for measurement and alarming is based on established physiological principles and programmed algorithms.

9. How the ground truth for the training set was established

Not applicable, as there is no mention of a machine learning "training set." The device's operation is based on known physical principles and sensor technology, not data-driven learning.

Ask a specific question about this device

MFM-CMS is a software application that is intended for use as a clinical data managing system (also referred to as a clinical information system - CIS).The MFM-CMS Central Monitoring System offers centralized physiological information management of adult, pediatric and neonatal patients which is automatically acquired from multiple bedside monitors. The MFM-CMS provides: collection, display and documentation of data from bedside monitors, viewing of patient physiologic data at remote locations and alarms when the results of the physiologic parameters exceed the user defined limit. It operates with off-the-shelf software. The system is intended for use in a hospital/clinical environment.

The MFM-CMS Central Monitoring System is a software production which runs on PC platform with Microsoft Windows XP or Microsoft Windows 7 operating system. Through specified EDAN protocol, one MFM-CMS can connect with multi-monitor manufactured by EDAN to collect patients' information and monitoring data such as physiological waveforms, physiological parameters and alarms. The MFM-CMS can also send bidirectional control instructions to bedside monitors to change patients' information, alarm limits, and conduct NIBP measurements. The bedside Patient Physiological Monitors have been cleared by FDA under K101539, K120144, K110922, K113623, K113653 and K120173 separately. The monitoring information collect by the MFM-CMS can be saved and printed. At the same time, the old records can be searched conveniently and quickly.

Here's an analysis of the provided text regarding the acceptance criteria and study for the EDAN Instruments MFM-CMS Central Monitoring System:

This device documentation does not contain the level of detail typically found in a study demonstrating performance against specific acceptance criteria for an AI/CADe device. The 510(k) summary focuses more on the software's functionality, its classification, and its substantial equivalence to a predicate device. It lacks quantifiable performance metrics against acceptance criteria.

Therefore, for many of your requested points, the answer will be "Not provided in the text."

1. A table of acceptance criteria and the reported device performance

Explanation: The document describes the system's features and intended use but does not define specific performance acceptance criteria (e.g., accuracy, sensitivity, specificity for particular physiological events like arrhythmia detection, or data display latency). Consequently, no reported device performance metrics against such criteria are provided.

2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)

• Sample Size for Test Set: Not provided in the text.
• Data Provenance: Not provided in the text.

Explanation: The document mentions "Verification and validation testing was done on MFM-CMS" and lists "Software testing," "Risk analysis," "Safety testing," and "Performance test" as quality assurance measures. However, it does not detail the nature of these "tests," nor does it specify any test datasets, their sizes, or their origin.

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)

• Number of Experts: Not provided in the text.
• Qualifications of Experts: Not provided in the text.

Explanation: Since no detailed performance study or test set is described, there's no mention of experts or their role in establishing ground truth.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set

• Adjudication Method: Not provided in the text.

Explanation: As no test set with expert ground truth is described, there's no mention of an adjudication method.

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: No, an MRMC comparative effectiveness study was not conducted or described.
• Effect Size of Human Improvement: Not applicable, as no such study was performed.

Explanation: The MFM-CMS is a central monitoring system for displaying physiological data and alarms from bedside monitors, not a diagnostic AI/CADe device designed to assist human readers in interpreting images or signals more effectively. Its primary function is data aggregation and display. Therefore, an MRMC study comparing human performance with and without AI assistance is not relevant to this type of device and was not conducted.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done

• Standalone Performance Study: Not described in the text in terms of quantifiable metrics.

Explanation: The "Performance test" mentioned is too vague to determine if it constituted a standalone performance study with specific metrics. The device's primary function is to process and display data, not necessarily to perform standalone diagnostic calculations that would have defined standalone accuracy metrics. It is a "software production" that runs on a PC and monitors data, and connectivity/display functionality would be the main "performance" aspects.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.)

• Type of Ground Truth: Not provided in the text.

Explanation: Without details on any specific performance tests, the method for establishing ground truth is not described.

8. The sample size for the training set

• Sample Size for Training Set: Not applicable / Not provided in the text.

Explanation: This device is a software application for data collection, display, and alarm management. It is not an AI/Machine Learning device in the sense that it requires a "training set" to learn to perform a diagnostic task. It relies on predefined protocols and inputs from connected monitors.

9. How the ground truth for the training set was established

• How Ground Truth for Training Set was Established: Not applicable / Not provided in the text.

Explanation: As it's not an AI/ML device requiring a training set, the concept of establishing ground truth for training is not relevant here.

Ask a specific question about this device