Search Results

QMAPP® is intended for use by professional healthcare providers for physiological/hemodynamic monitoring. The system may be used to display and analyze surface ECG (Electrocardiogram), respiration, invasive pressures, pulse oximetry (SpO2), End tidal CO2 (EtCO2), fractional flow reserve (FFR), non-invasive blood pressure (NiBP), surface body temperature, cardiac output and intra-cardiac ECG. QMAPP® provides also clinical data acquisition, medical image/data processing and analytical assessment. QMAPP® is intended for use in the areas of, but not limited to cardiology, cardiac catheterization, electrophysiology, radiology, invasive radiology. QMAPP® can be used standalone and in networked environments. The system is intended for patient/procedural data management, such as documentation, logging, reporting, trending, storing, reviewing, carrying out clinical calculations and exporting various representations of the acquired data. Data may also be acquired from and/or send to other devices, such as physiological monitoring system, information management systems, image acquisition/storage devices and other medical devices.

The QMAPP® system offers a complete physiological/hemodynamic monitoring and reporting system. The system is built from three units: an Amplifier, Live Monitoring CPU and Reporting CPU. The Amplifier Unit has various sensors connected with the patient, e.g. ECG, SpO2 and NiBP. The Amplifier Unit is connected to the Live Monitoring CPU via a dedicated Ethernet connection. The acquired patient information can be visualized on a Live Monitoring CPU. Typically located in the technical room. A software application executed on the Live Monitoring CPU can visualize the patient information. Also the Amplifier Unit can be controlled, i.e. most importantly, to set acquisition and filtering parameters for the different sensors, by the Live Monitoring CPU. Optionally the Monitoring unit can be connected via a dedicated Ethernet connection to a Reporting CPU, typically located in the technical room. On the Reporting CPU a database is installed which facilitates data storage and retrieval. A software application executed on the Reporting CPU serves as a patient data management system. It can e.g. be used for analysis, calculation and reporting in various representations of patient information.
The QMAPP® system, can operate standalone or it can be part of a typical hospital network infrastructure. The latter offers the possibility to send or receive information from and to other devices. The software has several communication modules, based on HL7 or DICOM protocols to interface with third party equipment/systems.
• The QMAPP® system works with 3rd party 510(k) cleared SpO2 module (Covidien Nellcor, K083325), NiBP module (CAS Medical Systems, MAXNIBP ND+, e.g. used in FDA cleared device CAS Medical Systems, 740 Select, K150620) and EtCO2 sensors e.g. used in FDA cleared device CLEO Patient Monitor, K142244.

The provided FDA 510(k) Clearance Letter for the QMAPP® System describes the device, its intended use, and a summary of non-clinical tests conducted to support its substantial equivalence. However, the document does not contain the specific details required to fully address your request regarding acceptance criteria and the comprehensive study that proves the device meets them.

Here's a breakdown of what can and cannot be extracted from the provided text, and where the requested information is missing:

Information Present in the Document:

• Overall Device Performance: The "NON-CLINICAL TESTS" section lists various characteristics on which "Bench testing" was carried out, implicitly suggesting these are areas where performance was evaluated. The "Referenced Standards and Performance Testing" section explicitly states that the QMAPP® system "meets the requirements of following performance Standards."
• Study Type: The studies mentioned are "Bench testing," "Usability Testing," and "Software verification and validation testing." The clearance is based on a "Traditional 510(k)" and relies on "non-clinical data."
• Ground Truth Type (for non-clinical testing): For the performance characteristics listed (ECG, Heart rate, SpO2, NiBP, IBP, Cardiac Output, Intra cardiac ECG, Skin Temperature, ECG impedance for Rate of respiratory effort, Measurement accuracy), the "ground truth" would be established by the physical standards and reference systems used during bench testing for each specific measurement. For example, a calibrated heart rate simulator would provide the ground truth for heart rate accuracy.
• Sample Size for Training Set: Not explicitly mentioned, but the document refers to a "software verification and validation testing," implying a dataset (likely synthetic or previously collected) was used.
• How Ground Truth for Training Set was Established: Not explicitly mentioned.

Missing Information (Crucial for your request):

The document focuses on demonstrating substantial equivalence to predicate devices through technical characteristics and adherence to recognized standards. It does not present a detailed study report with specific acceptance criteria, reported performance against those criteria, or the methodology of how "ground truth" was established for clinical or test datasets in the manner you've requested for an AI/ML context.

The QMAPP® system is a physiological/hemodynamic monitoring system, not specifically an AI/ML device that requires a comparison of algorithmic output against expert consensus on a test set, multi-reader multi-case studies, or standalone algorithm performance. The "clinical data acquisition, medical image/data processing and analytical assessment" mentioned are functions of the system, but the document does not elaborate on an AI/ML component with associated performance metrics.

Based on the provided text, here is what can be inferred and explicitly stated, with clear indications of missing information for your request:

1. Table of Acceptance Criteria and Reported Device Performance

The document states that the QMAPP® system was tested against and "meets the requirements of following performance Standards." These standards themselves contain detailed acceptance criteria for various parameters. The table below excerpts the performance characteristics mentioned in the "SUBSTANTIAL EQUIVALENCE SUMMARY TABLE" and "NON-CLINICAL TESTS" sections. Crucially, the document does not provide the specific numerical acceptance criteria (e.g., minimum accuracy percentages, maximum deviations) or the actual measured performance values against those criteria in a consolidated table. Instead, it states that the device "meets the requirements" of the listed standards and has "Accuracy" values which are the specifications for the device itself, not acceptance criteria of a study.

2. Sample size used for the test set and the data provenance

• Sample Size for Test Set: Not specified. The document only mentions "Bench testing," "Usability Testing," and "Software verification and validation testing." These are typically performed in a lab environment.
• Data Provenance (e.g., country of origin of the data, retrospective or prospective): Not specified. Given it's bench testing, actual patient data provenance is not directly relevant for the stated tests, but the data used for software verification and validation testing (if involving patient data) is not detailed.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts

• Number of Experts & Qualifications: Not applicable/not specified. For bench testing of physiological monitoring devices, the "ground truth" comes from calibrated testing equipment and reference signals, not expert human interpretation in the way, for example, a radiology AI would be evaluated. The "Software verification and validation testing" is also not described as relying on expert review of a patient dataset for ground truth.

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

• Adjudication Method: Not applicable/not specified. This methodology is typically used when comparing an algorithm's output to human expert interpretations, which is not the type of testing described for this physiological monitor.

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: Not applicable. The QMAPP® system is described as a physiological/hemodynamic monitoring, data acquisition, and analytical assessment system. It is not presented as an AI-assisted diagnostic tool designed to improve human reader performance in interpreting images or complex clinical scenarios.

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

• Standalone Performance: The described "Bench testing" and "Software verification and validation testing" can be considered "standalone" in the sense that they evaluate the device's inherent measurement and processing capabilities without a human in the loop for interpretation, but for a physiological monitor, the ultimate "human-in-the-loop" is the clinician using the displayed information. The document does not describe an AI algorithm that operates entirely independently to make a diagnosis or prediction in the same way an AI for image analysis might.

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

• Type of Ground Truth: For the "Bench testing" of physiological parameters, the ground truth would be established by calibrated reference standards and simulated physiological signals. For instance, a signal generator provides a known ECG waveform or blood pressure reading, and the device's measurement is compared to this known input.

8. The sample size for the training set

• Sample Size for Training Set: Not specified. The document mentions "Software verification and validation testing," which would involve a dataset, but its size is not detailed. There is no mention of a "training set" in the context of an AI/ML model, as the device is not presented as such.

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

• How Ground Truth for Training Set was Established: Not specified. If a "training set" was used for software validation (e.g., for signal processing algorithms), the ground truth would likely be established through
  • Synthetic data: Ground truth is known by design.
  • Previously validated physiological data: Data collected with highly accurate reference devices, where the "truth" for various physiological parameters is established by the reference device's measurements.

In summary: The FDA 510(k) clearance document for the QMAPP® System confirms that the device meets relevant performance standards through non-clinical bench testing and software validation. However, it does not provide the detailed acceptance criteria and study particulars, particularly those related to expert-adjudicated test sets, MRMC studies, or specific AI/ML training/testing methodologies, because the device is presented as a traditional physiological monitor, not an AI-powered diagnostic system.

Ask a specific question about this device

Fetal & Maternal Monitor (Model: F15A, F15A Air) is intended for providing Non-Stress testing or fetal monitoring for pregnant women from the 28th week of gestation. It is intended to be used only by trained and qualified personnel in antepartum examination rooms, labor and delivery rooms.

Fetal & Maternal Monitor (Model: F15A, F15A Air) is intended for real time monitoring of fetal and maternal physiological parameters, including non-invasive monitoring and invasive monitoring:

Non-invasive physiological parameters:

• Maternal heart rates (MHR)
• Maternal ECG (MECG)
• Maternal temperature (TEMP)
• Maternal oxygen saturation (SpO2) and pulse rates (PR)
• Fetal heart rates (FHR)
• Fetal movements (FM)
• FTS-3

Note: SpO2 and PR are not available in F15A Air.

Invasive physiological parameters:

• Uterine activity
• Direct ECG (DECG)

The F15A series fetal and maternal monitor can monitor multiple physiological parameters of the fetus/mother in real time. F15A series can display, store, and print patient information and parameters, provide alarms of fetal and maternal parameters, and transmit patient data and parameters to Central Monitoring System.

F15A series fetal and maternal monitors mainly provide following primary feature:

Non-invasive physiological parameters:

• Maternal heart rates (MHR)
• Maternal ECG (MECG)
• Maternal temperature (TEMP)
• Maternal oxygen saturation (SpO2) and pulse rates (PR)
• Fetal heart rates (FHR)
• Fetal movements (FM)
• FTS-3

Note: SpO2 and PR are not available in F15A Air.

Invasive physiological parameters:

• Uterine activity
• Direct ECG (DECG)

The provided FDA 510(k) clearance letter and summary for the Fetal & Maternal Monitor (F15A, F15A Air) do not contain the detailed information necessary to fully answer all aspects of your request regarding acceptance criteria and the study that proves the device meets them.

The document focuses primarily on demonstrating substantial equivalence to a predicate device (Edan Instruments, Inc., F9 Express Fetal & Maternal Monitor, K173042) through comparison of intended use, technological characteristics, and conformance to various safety and performance standards. It mentions "functional and system level testing to validate the performance of the devices" and "results of the bench testing show that the subject device meets relevant consensus standards," but it does not specify quantitative acceptance criteria for each individual physiological parameter (e.g., FHR accuracy, SpO2 accuracy) nor the specific results of those tests beyond stating that they comply with standards.

Specifically, the document does not include information on:

• A table of acceptance criteria with specific quantitative targets for each parameter and the reported device performance values against those targets. It only states compliance with standards.
• Sample sizes used for a "test set" in the context of clinical performance evaluation (it mentions "bench testing," but this is typically laboratory-based and doesn't involve patient data in a "test set" sense for AI/algorithm performance validation).
• Data provenance for such a test set (e.g., country of origin, retrospective/prospective).
• Number or qualifications of experts used to establish ground truth.
• Adjudication methods.
• Multi-Reader Multi-Case (MRMC) studies or human reader improvement data with AI assistance.
• Standalone (algorithm-only) performance, as this is a monitoring device, not a diagnostic AI algorithm.
• Type of ground truth (beyond "bench testing" which implies engineered signals or controlled environments).
• Sample size for a training set or how ground truth for a training set was established. This device is a traditional medical device, not an AI/ML-driven diagnostic or interpretative algorithm in the way your request implies.

Therefore, based solely on the provided text, I can only address what is present or infer what is missing.

Here's a breakdown based on the available information:

Analysis of Acceptance Criteria and Performance Testing based on Provided Document

The provided 510(k) summary focuses on demonstrating substantial equivalence to a predicate device (F9 Express Fetal & Maternal Monitor, K173042) by showing that the new device (F15A, F15A Air) has the same intended use and fundamentally similar technological characteristics, with any differences not raising new safety or effectiveness concerns.

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

The document does not provide a specific table with quantitative acceptance criteria for each physiological parameter (e.g., FHR accuracy, SpO2 accuracy) and the corresponding reported performance values obtained in testing. Instead, it states that the device was assessed for conformity with relevant consensus standards. For example, it lists:

• IEC 60601-2-37:2015: Particular requirements for the basic safety and essential performance of ultrasonic medical diagnostic and monitoring equipment (relevant for FHR).
• ISO 80601-2-61:2017+A1:2018: Particular requirements for basic safety and essential performance of pulse oximeter equipment (relevant for SpO2).
• ISO 80601-2-56:2017+A1:2018: Particular requirements for basic safety and essential performance of clinical thermometers for body temperature measurement (relevant for TEMP).
• IEC 60601-2-27:2011: Particular requirements for the basic safety and essential performance of electrocardiographic monitoring equipment (relevant for MECG/DECG).

Acceptance Criteria (Inferred from standards compliance): The acceptance criteria are implicitly the performance requirements specified within these listed consensus standards. These standards set limits for accuracy, precision, response time, and other performance metrics for each type of measurement.

Reported Device Performance: The document states: "The results of the bench testing show that the subject device meets relevant consensus standards." This implies that the measured performance statistics (e.g., accuracy, bias, precision) for each parameter fell within the acceptable limits defined by the respective standards. However, the specific measured values are not provided in this summary.

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

The document mentions "Bench Testing" which implies laboratory-based testing using simulators, controlled signals, or phantoms, rather than a "test set" involving patient data. There is no information provided regarding:

• Sample size (e.g., number of recordings, duration of recordings, number of simulated cases) for the bench tests for each parameter.
• Data provenance (e.g., country of origin, retrospective or prospective) as this is not a study involving patient data collection for performance validation.

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 and not provided. For a traditional physiological monitor, ground truth for bench testing is typically established using:

• Calibrated reference equipment/simulators: e.g., ECG simulators to generate known heart rates, SpO2 simulators to generate known oxygen saturation levels.
• Physical standards/phantoms: e.g., temperature baths at known temperatures.
• Known physical properties: e.g., precise weights for pressure transducers.

Clinical experts are not involved in establishing ground truth for bench performance of these types of physiological measurements.

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

This is not applicable and not provided. Adjudication methods are relevant for human expert review of complex clinical data (e.g., medical images for AI validation) to establish a consensus ground truth. For bench testing of physiological monitors, ground truth is objectively determined by calibrated instruments or defined physical parameters.

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

This is not applicable and not provided. An MRMC study is typically performed to evaluate the diagnostic accuracy of AI-assisted human interpretations versus unassisted human interpretations for AI-driven diagnostic devices. The Fetal & Maternal Monitor is a physiological monitoring device, not an AI-assisted diagnostic imaging or interpretation system. It measures and displays physiological parameters; it does not provide AI-driven assistance for human "readers" to interpret complex clinical information.

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

The device is a monitor that directly measures physiological parameters. It is not an "algorithm only" device in the sense of an AI model providing a diagnostic output. Its performance (e.g., FHR accuracy) is its standalone performance, as it directly measures these parameters. The document states "functional and system level testing to validate the performance of the devices," which would represent this type of standalone performance for the measurement functionalities. However, specific quantitative results are not given, only compliance with standards.

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

As explained in point 3, the ground truth for bench testing of physiological monitors is established using calibrated reference equipment/simulators and physical standards.

8. The sample size for the training set

This is not applicable and not provided. This device is a traditional physiological monitor, not a machine learning model that requires a "training set." Its algorithms for parameter measurement are based on established physiological principles and signal processing techniques, not on statistical learning from large datasets.

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

This is not applicable and not provided for the same reasons as point 8.

Ask a specific question about this device

The indications for use of the BeneVision Central Monitoring System include:

• Real time viewing of patient clinical data and alarms from compatible physiological monitors. Viewing of non-real time patient clinical data of compatible anesthesia devices (i.e. not indicated for real-time monitoring of clinical data of compatible anesthesia devices).

• Storage and Historical review of patient clinical data and alarms from compatible physiological monitor, and anesthesia devices.

• Printing patient data from compatible physiological monitor, and anesthesia devices.

• Configuration of local settings as well as synchronizing settings across the network to remote compatible physiological monitors.

• Transfer of patient clinical data and settings between several CentralStations.

• Provides a Resting 12 Lead interpretation of previously stored data.

The BeneVision Central Monitoring System is a networked patient monitoring system intended for use in a fixed location, installed in professional healthcare facilities to provide clinicians remote patient monitoring. The network connections between the various devices can be any combination of Ethernet (Wired), Wireless WIFI (WLAN), and Wireless WMTS.

The BeneVision Central Monitoring System supports one or more Mindray compatible physiological monitors, anesthesia systems and will display, store, print, and transfer information received from the compatible monitors, anesthesia systems.

The telemetry monitoring systems are designed to acquire and monitor physiological data for ambulating patients within a defined coverage area. The BeneVision Central Monitoring System supports Telemetry Systems: TMS-6016, Telepack-608, TMS60, TM80, and TM70.

• The TMS-6016 transmitter is intended for use on Adult and Pediatric patients to monitor ECG and SpO2 physiological data.

• The Panorama Telepack-608 transmitter is intended for use on Adult patients to monitor ECG and SpO2 physiological data.

• The TMS60 transmitter is intended for use on Adult and Pediatric patients over three years old to monitor ECG, SpO2, NIBP and Resp physiological data. The physiological data can be reviewed locally on the display of the transmitter. The CentralStation will support ECG, Heart Rate, SpO2, NIBP, Resp, Pulse Rate, Arrhythmia analysis, QT monitoring, and ST Segment Analysis for the TMS60.

• The TM80/TM70 telemetry monitor is intended for use on Adult and Pediatric patients over three years old to monitor ECG, SpO2, NIBP and Resp physiological data. The physiological data can be analyzed, alarmed, stored, reviewed locally on the display of the monitor, and the CentralStation can config and display the physiological parameters from the TM80/TM70.

The BeneVision Central Monitoring System is intended for use in professional healthcare facilities under the direct supervision of a licensed healthcare practitioner.

The BeneVision Central Monitoring System (CMS) is a networked patient monitoring system intended for use in healthcare settings by, or under the direction of, a physician to provide clinicians remote patient monitoring. The target patient population is adult patients and pediatrics.

When connected to a compatible anesthesia device, BeneVision CMS can display the parameters, waveforms and alarms of the anesthesia device. The device does not contain bi-directional capabilities for the compatible anesthesia devices.

The BeneVision CMS includes the AlarmGUARD application. AlarmGUARD supports delivering notifications of physiological and technical alarms to clinical professionals' mobile devices. AlarmGUARD is not intended for real time monitoring of patients and is not intended to act as a primary source for alarms.

It appears the provided FDA 510(k) clearance letter and summary for the BeneVision Central Monitoring System (K242728) does not contain specific acceptance criteria, test results (like sensitivity/specificity, accuracy metrics), or detailed study methodologies that directly address how the device's performance meets quantitative acceptance criteria for its intended functions.

The document primarily focuses on demonstrating substantial equivalence to a predicate device (K220058) through:

• Comparison of Indications for Use: Showing minor differences (expanded compatibility to include anesthesia systems, but not for real-time monitoring).
• Technological Comparisons: Highlighting changes in operating systems, host configurations, and the addition of features like Multi-Patient Viewer separation and AlarmGUARD support.
• Performance Data Section: This section lists the types of testing conducted but does not provide the results of those tests or specific acceptance criteria met by those results. It merely states that "Software verification and validation testing was conducted" and "Verification of the BeneVision Central Monitoring System was conducted to ensure that the product works as designed. Validation was conducted to check the design and performance of the product."

Therefore, based solely on the provided text, I cannot extract the detailed information requested in your prompt regarding acceptance criteria, reported performance, sample sizes, expert qualifications, adjudication methods, MRMC studies, standalone performance, or training set details.

The document confirms the following regarding the study:

• Study Type: Software verification and validation testing, along with specific bench testing.
• Clinical Data/Animal Testing: Not applicable/not required for this submission to establish substantial equivalence. This suggests the clearance relies on non-clinical data and comparison to the predicate.
• Ground Truth: The document implies that the ground truth for software verification and validation would be the design specifications and expected behavior of the system, rather than clinical outcomes or expert consensus on a diagnostic task. For the "Waveform Display Accuracy from compatible Anesthesia Machine," the ground truth would likely be the direct output from the anesthesia machine itself.

What is present in the document that somewhat relates to your request:

• "Bench Testing" section (Page 19): This lists specific tests performed:
  • AlarmGUARD IEC 60601-2-27
  • AlarmGUARD IEC 60601-1-8
  • AlarmGUARD Human Factors
  • Waveform Display Accuracy from compatible Anesthesia Machine

To fulfill your request as best as possible with the given information, I will have to state that many details are explicitly absent from this public 510(k) summary.

Here's a structured response based on the provided document, indicating what information is present and what is absent:

Device Acceptance Criteria and Performance Study Summary (K242728)

Based on the provided FDA 510(k) Clearance Letter and Summary, detailed quantitative acceptance criteria and specific performance metrics (like accuracy, sensitivity, specificity) for the BeneVision Central Monitoring System are not explicitly presented. The submission primarily relies on demonstrating substantial equivalence to a predicate device (K220058) through verification and validation of software and specific bench testing.

The document states that "Software verification and validation testing was conducted and documentation was provided as recommended by FDA's Guidance 'Content of Premarket Submissions for Device Software Functions: Guidance for Industry and Food and Drug Administration Staff'." It also mentions that "Verification of the BeneVision Central Monitoring System was conducted to ensure that the product works as designed. Validation was conducted to check the design and performance of the product."

1. Table of Acceptance Criteria and Reported Device Performance

2. Sample Size Used for the Test Set and Data Provenance

• Test Set Sample Size: Not specified in the provided document for any of the listed tests (AlarmGUARD, Waveform Display Accuracy, general software V&V).
• Data Provenance: Not specified (e.g., country of origin, retrospective/prospective). Given that no clinical data was used or required, the "data" would be synthetic, simulated, or derived from direct device connections during bench testing.

3. Number of Experts and Qualifications for Ground Truth

• Not applicable / Not specified. The document does not describe the use of human experts to establish ground truth for a diagnostic task or for the performance evaluation of this central monitoring system. The focus is on software function and electro-mechanical performance validation against design specifications and international standards.

4. Adjudication Method for the Test Set

• Not applicable / Not specified. No adjudication method is mentioned as human reader input for a test set is not described.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• No. The document explicitly states that "Clinical testing is not required to establish substantial equivalence to the predicate device" and does not mention any MRMC study. This device is a central monitoring system displaying physiological data, not an AI diagnostic tool requiring MRMC studies for improved human reader performance.

6. Standalone Performance (Algorithm Only)

• The "performance data" section lists "Software Verification and Validation Testing" and "Bench Testing" (including "Waveform Display Accuracy from compatible Anesthesia Machine"). These tests conceptually represent 'standalone' performance in that they evaluate the device's technical functions directly. However, no specific quantitative standalone performance metrics (e.g., classification accuracy, sensitivity, specificity for any internal algorithms) are provided in this summary beyond the statement that v&v was conducted to ensure the product "works as designed."

7. Type of Ground Truth Used

• The ground truth for the device's performance appears to be:
  • Design Specifications: For general software verification and validation.
  • External Reference Standards/Simulators: For tests like "Waveform Display Accuracy" (e.g., comparing the displayed waveform to the known, true signal generated by a simulator or the anesthesia machine itself).
  • International Standards: For AlarmGUARD functionality (e.g., IEC 60601-2-27, IEC 60601-1-8).

8. The Sample Size for the Training Set

• Not applicable / Not specified. This document describes a traditional medical device (patient monitoring system software) rather than a machine learning/AI device that typically requires a distinct "training set." Therefore, no training set size is mentioned.

9. How the Ground Truth for the Training Set Was Established

• Not applicable / Not specified. As no training set for an AI/ML model is indicated, there is no mention of how its ground truth would be established.

Ask a specific question about this device

Sleepiz One+ is a contactless medical device intended to measure heart rate and respiration rate in adult patients, at rest or during sleep (in non-motion condition) and to detect patient presence and body movements.

The Sleepiz One+ hardware unit is intended to be used by a healthcare professional when the recordings are performed in a clinical setting, or by patients or their caregivers when the recordings are performed in a home environment. The Sleepiz One+ cloud software is intended for use by healthcare professionals.

This device is not indicated for active patient monitoring, as it does not provide alarms for timely response in life-threatening situations. It is not indicated for use on pregnant women or patients with active implantable devices.

Sleepiz One+ is a contactless medical device that uses radar technology to measure respiration rate and heart rate from a resting or sleeping patient.

The Sleepiz One+ consists of a hardware unit and cloud-based software The hardware unit can be positioned on a bedside table, mounted on a stand, or attached to the wall behind the patient's bed. It is designed to monitor physiological signals by detecting small body movements, such as those caused by breathing and heartbeat, using Doppler radar. The recorded signals are then transmitted via Wi-Fi to cloud-based software, where they are analyzed to obtain respiration rate, heart rate, and body movement. These outputs can be exposed via the Application Programming Interfaces (APIs) to allow healthcare professionals the review and annotation of the data and compilation of results into reports.

Outputs

• Breathing pattern
• Instantaneous breathing rate [breaths per minute]
• Breathing rate statistics (10th, 50th, and 90th quantiles) [breaths per minute]
• Body movement
• Time in bed [hours]
• Presence detection
• Heart rate [beats per minute]
• Heart rate statistics (10th, 50th, and 90th quantiles) [beats per minute]

The overall system can be grouped into 4 major components, which are classified on the basis of the logical component interfaces where data exchange is occurring.

• Sleepiz Hardware - This is a hardware component serving as primary data acquisition device.
• Embedded Software - This encompasses the firmware running of the Sleepiz Hardware. This, together with Sleepiz hardware, is responsible for data acquisition. The embedded software forms the crux of the Sleepiz hardware such that it defines and controls the data acquisition process. The security aspects related to the operation of the device are incorporated in the design and implementation of embedded software.
• Sleep Analytics Software - The sleep analytics software is responsible for processing data from the Sleepiz Hardware and returning its analytics (e.g., breathing rate, heart rate), as well as its statistics (e.g., mean breathing rate, total recording time, etc.). This refers to the ML model deployed within the cloud software. By itself, the Sleep Analytics Software does not have an external interface. It is wholly encapsulated by the cloud software component Data Processing Layer.
• Cloud Software - The cloud software can be divided into the backend service and the analytics service. The backend service includes modules for data ingestion, a public API, a private API, and a module for sending analysis process requests. The analytics service is responsible for receiving analysis requests and interacting with the sleep analytics software.

The FDA 510(k) clearance letter for Sleepiz One+ (2.5) indicates that the device's substantial equivalence to a predicate device (Sleepiz One+) was established primarily through non-clinical performance testing, focusing on software verification and validation, electrical safety, and electromagnetic compatibility. The document states that the subject device and the predicate device have the same intended use, principle of operation, and similar technological characteristics. The minor modifications (API-based data access instead of web interface and plug-in power instead of battery) were assessed for safety and effectiveness without requiring extensive new clinical studies.

The provided document does not detail specific acceptance criteria for the accuracy of heart rate and respiration rate measurements, nor does it provide a study proving the device meets these specific performance criteria. The clearance is based on the conclusion that "The verification and validation tests performed on the subject device confirm that the device performs as intended in the specified use conditions and comparably to the predicate device. The performance testing conducted shows comparable results between the two models, thereby, demonstrating the safety and performance of the Sleepiz One+ (V.2.5)."

Therefore, based solely on the provided text, a comprehensive table of acceptance criteria and the detailed study proving the device's accuracy against those criteria cannot be constructed. The document infers that the device performs as intended and comparably to the predicate, but it does not present the raw performance data or the specific acceptance thresholds for heart rate and respiration rate accuracy.

Here's a breakdown of the requested information based on the provided text, with clear indications where the information is NOT available.

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

Information NOT available in the provided text. The 510(k) summary states that "performance testing conducted shows comparable results between the two models," but it does not quantify these results or list specific acceptance criteria for heart rate and respiration rate accuracy. The testing primarily focused on software validation, electrical safety, and EMC.

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

Information NOT available in the provided text. The document mentions "non-clinical performance tests" and "verification and validation tests," but it doesn't specify any sample sizes for a test set related to the accuracy of vital sign measurements, nor the provenance of any data used for such testing.

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)

Information NOT available in the provided text. There is no mention of external experts or ground truth establishment in the context of vital sign accuracy, as the clearance seems to rely on comparability to a predicate device and engineering verification/validation. For devices measuring physiological parameters, ground truth is typically established via reference medical devices (e.g., ECG, capnography) rather than expert consensus on subjective data.

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

Information NOT available in the provided text. Since no details on a clinical or performance study involving human subjects and expert review are provided, there is 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

Information NOT available in the provided text. The Sleepiz One+ is a vital signs monitor, not typically an AI-assisted diagnostic imaging device that would undergo MRMC studies. The device primarily measures heart rate and respiration rate via radar and processes this data in cloud software (Sleep Analytics Software contains the ML model). The interaction is between the device and the patient, not a "human reader" interpreting AI outputs in a MRMC context. The cleared device provides raw data and statistics (like quantiles), which healthcare professionals would then interpret.

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

The device's core function, measuring heart rate and respiration rate from radar signals using its Sleep Analytics Software (which contains the ML model), operates in a standalone manner to generate these outputs. The outputs themselves (instantaneous rates, statistics) are generated by the algorithm without human intervention in the loop of the measurement process itself. Healthcare professionals then access and interpret these results via API. The non-clinical performance testing would have validated the output of this standalone algorithm against internal specifications or a reference.

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

Information NOT explicitly stated in the provided text regarding vital sign accuracy. For devices measuring heart rate and respiration rate, ground truth is typically established using established reference medical devices (e.g., synchronously recorded ECG for heart rate, capnography or impedance pneumography for respiration rate). While not stated, it can be inferred that if accuracy was evaluated, it would be against such objective physiological measurements rather than subjective expert consensus or pathology.

8. The sample size for the training set

The device uses an "ML model deployed within the cloud software" for "Sleep Analytics Software." However, the provided 510(k) summary does NOT provide any details about the training set size or methodology for this ML model. The focus of the substantial equivalence claim is on the overall system's safety and performance comparability to the predicate, with modifications primarily linked to data access and power source.

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

Information NOT available in the provided text. As with the training set size, the 510(k) summary does not provide details on how the ground truth for the training set (if any specific to the ML model) was established.

Ask a specific question about this device

The monitors are intended to be used for monitoring, storing, recording, and reviewing of, and to generate alarms for, multiple physiological parameters of adults and pediatrics (including neonates). The monitors are intended for use by trained healthcare professionals in hospital environments.

The monitored physiological parameters include: ECG, respiration (RESP), temperature (TEMP), functional oxygen saturation of arterial hemoglobin (SpO₂), pulse rate (PR), non-invasive blood pressure (NIBP), invasive blood pressure (IBP), carbon dioxide (CO2), and cardiac output (C.O.).

The arrhythmia detection and ST Segment analysis are intended for adult patients.

The NIBP monitoring supports iCUFS algorithm and iFAST algorithm. The iCUFS algorithm is intended for adult, pediatric and neonatal patients. The iFAST algorithm is intended for adult and pediatric patients (≥3 years of age). Both measurement algorithms are also intended for use with pregnant women, including pre-eclamptic patients. NIBP MAP is not applicable to pregnant women.

The Spot Temp with T2A module can only measure temperature of adult and pediatric (> 1 year of age) patients.

The monitors are not intended for MRI environments.

The cardiac output (C.O.) is only intended for adult patients.

The CX&UX series Patient Monitor including CX10/CX12/CX15/UX10/UX12/UX15 can perform long-time continuous monitoring of multiple physiological parameters. Also, it is capable of storing, displaying, analyzing and controlling measurements, and it will indicate alarms in case of abnormalities so that doctors and nurses can respond to the patient's situation as appropriate.

Minor differences from the predicate device are limited to some modifications of monitoring parameter specifications. These updates do not change the fundamental scientific technology of the cleared predicate device and thus do not raise any questions about the safety and effectiveness of the subject device.

The provided FDA 510(k) clearance letter details the device's technical specifications and comparisons to predicate devices, along with the non-clinical performance data and adherence to various IEC and ISO standards. However, it explicitly states: "Clinical data: The subject device did not require new clinical studies to support substantial equivalence."

This means that the submission for this Patient Monitor device (CX10, CX12, CX15, UX10, UX12, UX15) relies on demonstrating substantial equivalence to a legally marketed predicate device (Edan Instruments, Inc., Patient Monitor Model iX10, iX12, iX15, K232962) through non-clinical performance testing and software verification/validation, rather than new clinical trials or studies involving human patients.

Therefore, the requested information regarding acceptance criteria and studies that prove the device meets acceptance criteria through clinical performance (e.g., sample size for test set, expert involvement, MRMC studies, ground truth establishment for test/training sets, effect size of human reader improvement with AI) cannot be extracted from this document, as such clinical studies were explicitly not required for this 510(k) submission.

The document focuses on demonstrating that the new device's technical specifications and performance are similar to the predicate device, and that it complies with relevant safety and performance standards through bench testing.

Here's what can be extracted from the provided text regarding acceptance criteria and the type of study performed, specifically focusing on the non-clinical aspects:

Device: Patient Monitor (CX10, CX12, CX15, UX10, UX12, UX15)

The acceptance criteria for this device are implicitly tied to its performance meeting the standards and accuracy specifications of the predicate device and relevant international standards. Since no new clinical studies were conducted, the "proof" comes from non-clinical bench testing and software validation.

1. Table of Acceptance Criteria and Reported Device Performance (Non-Clinical/Bench Testing)

2. Sample Size Used for the Test Set and Data Provenance:

• Sample Size: Not applicable in terms of human subjects or patient data test sets, as "new clinical studies" were not required. The "test set" refers to bench testing and functional system-level validation. The specific number of test cycles or a detailed breakdown of test cases for bench testing is not provided in this summary.
• Data Provenance: The data primarily originates from Edan Instruments Inc. (Shenzhen, Guangdong, China) through internal engineering and quality assurance processes for non-clinical bench testing and software validation. It is not patient data, so concepts like "retrospective or prospective" do not apply.

3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications:

• Not applicable for clinical ground truth: Since no clinical studies were performed requiring human interpretation or diagnosis for a test set, no medical experts (e.g., radiologists) were used to establish ground truth in this context.
• Internal experts: Bench testing and software validation would have involved engineers and quality assurance professionals, whose qualifications are implicit in the quality system (21 CFR Part 820) but not specified in detail here.

4. Adjudication Method for the Test Set:

• Not applicable: Adjudication methods (e.g., 2+1, 3+1) are relevant for clinical studies involving multiple readers. This was not a clinical study. Bench testing relies on established technical specifications and standard compliance.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was Done:

• No: No MRMC study was performed as no new clinical studies were required or conducted. Therefore, there's no effect size of human readers improving with AI assistance. The device is a patient monitor, not an AI-assisted diagnostic tool.

6. If a Standalone (i.e. algorithm only without human-in-the loop performance) was Done:

• Yes (for the technical components): The "performance testing-Bench" effectively represents a standalone evaluation of the device's functional components (ECG, NIBP, SpO2, etc.) and software against defined technical specifications and standards. The "software verification and validation testing" also represents a standalone evaluation of the algorithm and software functions. The specific algorithms (e.g., iCUFS, iFAST for NIBP, arrhythmia analysis logic) are tested independently for their accuracy against known inputs or reference standards as part of bench testing.

7. The Type of Ground Truth Used:

• Technical/Reference Standards: For the bench testing, the "ground truth" would be derived from:
  • Reference standards/simulators: Calibrated medical equipment, physiological simulators, and test signals (e.g., known ECG waveforms, simulated blood pressure readings, temperature standards) are used to provide the "true" values against which the device's measurements are compared.
  • Defined specifications: The device's internal design specifications and the requirements of the referenced IEC/ISO standards serve as the "ground truth" for compliance testing.
• Not clinical ground truth: No expert consensus, pathology, or outcomes data from real patients were used for establishing ground truth for this submission.

8. The Sample Size for the Training Set:

• Not applicable: The device is a patient monitor, not a machine learning/AI algorithm that typically undergoes a distinct "training" phase with a large dataset. Its functionality is based on established physiological measurement principles and programmed algorithms. Any internal calibration or algorithm refinement would be part of the product development process, not a dedicated "training set" in the AI/ML sense.

9. How the Ground Truth for the Training Set Was Established:

• Not applicable: As there was no "training set" in the context of an AI/ML model, the concept of establishing ground truth for it does not apply to this 510(k) submission.

In summary, this 510(k) clearance relies on demonstrating that the new Patient Monitor is substantially equivalent to a previously cleared predicate device, primarily through robust non-clinical bench testing and software validation, proving compliance with established medical device standards and functional specifications. No new clinical studies with patient data were required or conducted for this specific submission.

Ask a specific question about this device

The monitor B105M, B125M, B155M, B105P and B125P are portable multi-parameter patient monitors intended to be used for monitoring, recording, and to generate alarms for multiple physiological parameters of adult, pediatric, and neonatal patients in a hospital environment and during intra-hospital transport.

The monitor B105M, B125M, B155M, B105P and B125P are intended for use under the direct supervision of a licensed health care practitioner.

The monitor B105M, B125M, B155M, B105P and B125P are not Apnea monitors (i.e., do not rely on the device for detection or alarm for the cessation of breathing). These devices should not be used for life sustaining/supporting purposes.

The monitor B105M, B125M, B155M, B105P and B125P are not intended for use during MRI.

The monitor B105M, B125M, B155M, B105P and B125P can be stand-alone monitors or interfaced to other devices via network.

The monitor B105M, B125M, B155M, B105P and B125P monitor and display: ECG (including ST segment, arrhythmia detection, ECG diagnostic analysis and measurement), invasive blood pressure, heart/pulse rate, oscillometric non-invasive blood pressure (systolic, diastolic and mean arterial pressure), functional oxygen saturation (SpO2) and pulse rate via continuous monitoring (including monitoring during conditions of clinical patient motion or low perfusion), temperature with a reusable or disposable electronic thermometer for continual monitoring Esophageal/Nasopharyngeal/Tympanic/Rectal/Bladder/Axillary/Skin/Airway/Room/Myocardial/Core/Surface temperature, impedance respiration, respiration rate, airway gases (CO2, O2, N2O, anesthetic agents, anesthetic agent identification and respiratory rate), Cardiac Output (C.O.), Entropy, neuromuscular transmission (NMT) and Bispectral Index (BIS).

The monitor B105M, B125M, B155M, B105P and B125P are able to detect and generate alarms for ECG arrhythmias: Asystole, Ventricular tachycardia, VT>2, Ventricular Bradycardia, Accelerated Ventricular Rhythm, Ventricular Couplet, Bigeminy, Trigeminy, "R on T", Tachycardia, Bradycardia, Pause, Atrial Fibrillation, Irregular, Multifocal PVCs, Missing Beat, SV Tachy, Premature Ventricular Contraction (PVC), Supra Ventricular Contraction (SVC) and Ventricular fibrillation.

The proposed monitors B105M, B125M, B155M, B105P and B125P are new version of multi-parameter patient monitors developed based on the predicate monitors B105M, B125M, B155M, B105P and B125P (K213490) to provide additional monitored parameter Bispectral Index (BIS) by supporting the additional optional E-BIS module (K052145) which used in conjunction with Covidien BISx module (K072286).

In addition to the added parameter, the proposed monitors also offer below several enhancements:

• Provided data connection with GE HealthCare anesthesia devices to display the parameters measured from anesthesia devices (Applicable for B105M, B125M and B155M).
• Modified Early Warning Score calculation provided.
• Separated low priority alarms user configurable settings from the combined High/Medium/Low priority options.
• Provided additional customized notification tool to allow clinician to configure the specific notification condition of one or more physiological parameters measured by the monitor. (Applicable for B105M, B125M and B155M).
• Enhanced User Interface in Neuromuscular Transmission (NMT), Respiration Rate and alarm overview.
• Provided Venous Stasis to assist venous catheterization with NIBP cuff inflation.
• Supported alarm light brightness adjustment.
• Supported alarm audio pause by gesture (Not applicable for B105M and B105P).
• Supported automatic screen brightness adjustment.
• Supported network laser printing.
• Continuous improvements in cybersecurity

The proposed monitors B105M, B125M, B155M, B105P and B125P retain equivalent hardware design based on the predicate monitors and removal of the device Trim-knob to better support cleaning and disinfecting while maintaining the same primary function and operation.

Same as the predicate device, the five models (B105M, B125M, B155M, B105P and B125P) share the same hardware platform and software platform to support the data acquisition and algorithm modules. The differences between them are the LCD screen size and configuration options. There is no change from the predicate in the display size.

As with the predicate monitors B105M, B125M, B155M, B105P and B125P (K213490), the proposed monitors B105M, B125M, B155M, B105P and B125P are multi-parameter patient monitors, utilizing an LCD display and pre-configuration basic parameters: ECG, RESP, NIBP, IBP, TEMP, SpO2, and optional parameters which include CO2 and Gas parameters provided by the E-MiniC module (K052582), CARESCAPE Respiratory modules E-sCO and E-sCAiO (K171028), Airway Gas Option module N-CAiO (K151063), Entropy parameter provided by the E-Entropy module (K150298), Cardiac Output parameter provided by the E-COP module (K052976), Neuromuscular Transmission (NMT) parameter provided by E-NMT module (K051635) and thermal recorder B1X5-REC.

The proposed monitors B105M, B125M, B155M, B105P and B125P are not Apnea monitors (i.e., do not rely on the device for detection or alarm for the cessation of breathing). These devices should not be used for life sustaining/supporting purposes. Do not attempt to use these devices to detect sleep apnea.

As with the predicate monitors B105M, B125M, B155M, B105P and B125P (K213490), the proposed monitors B105M, B125M, B155M, B105P and B125P also can interface with a variety of existing central station systems via a cabled or wireless network which implemented with identical integrated WiFi module. (WiFi feature is disabled in B125P/B105P).

Moreover, same as the predicate monitors B105M, B125M, B155M, B105P and B125P (K213490), the proposed monitors B105M, B125M, B155M, B105P and B125P include features and subsystems that are optional or configurable, and it can be mounted in a variety of ways (e.g., shelf, countertop, table, wall, pole, or head/foot board) using existing mounting accessories.

The provided FDA 510(k) clearance letter and summary for K242562 (Monitor B105M, Monitor B125M, Monitor B155M, Monitor B105P, Monitor B125P) do not contain information about specific acceptance criteria, reported device performance metrics, or details of a study meeting those criteria for any of the listed physiological parameters or functionalities (e.g., ECG or arrhythmia detection).

Instead, the documentation primarily focuses on demonstrating substantial equivalence to a predicate device (K213490) by comparing features, technology, and compliance with various recognized standards and guidance documents for safety, EMC, software, human factors, and cybersecurity.

The summary explicitly states: "The subject of this premarket submission, the proposed monitors B105M/B125M/B155M/B105P/B125P did not require clinical studies to support substantial equivalence." This implies that the changes introduced in the new device versions were not considered significant enough to warrant new clinical performance studies or specific quantitative efficacy/accuracy acceptance criteria beyond what is covered by the referenced consensus standards.

Therefore, I cannot provide the requested information from the given text:

1. A table of acceptance criteria and the reported device performance: This information is not present. The document lists numerous standards and tests performed, but not specific performance metrics or acceptance thresholds.
2. Sample size used for the test set and the data provenance: Not explicitly stated for performance evaluation, as clinical studies were not required. The usability testing mentioned a sample size of 16 US clinical users, but this is for human factors, not device performance.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not applicable, as detailed performance studies requiring expert ground truth are not described.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set: Not applicable.
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 patient monitor, not an AI-assisted diagnostic tool that would typically involve human readers.
6. If a standalone (i.e. algorithm only without human-in-the loop performance) was done: The document describes "Bench testing related to software, hardware and performance including applicable consensus standards," which implies standalone testing against known specifications or simulated data. However, specific results or detailed methodologies for this type of testing are not provided beyond the list of standards.
7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.): Not explicitly stated for performance assessment. For the various parameters (ECG, NIBP, SpO2, etc.), it would typically involve reference equipment or validated methods as per the relevant IEC/ISO standards mentioned.
8. The sample size for the training set: Not applicable, as this is not an AI/ML device that would require explicit training data in the context of this submission.
9. How the ground truth for the training set was established: Not applicable.

In summary, the provided document focuses on demonstrating that the new monitors are substantially equivalent to their predicate through feature comparison, adherence to recognized standards, and various non-clinical bench tests (e.g., hardware, alarms, EMC, environmental, reprocessing, human factors, software, cybersecurity). It does not contain the detailed performance study results and acceptance criteria typically found for novel diagnostic algorithms or AI-driven devices.

Ask a specific question about this device

The Empatica Health Monitoring Platform is a wearable device and paired mobile and cloud-based software platform intended to be used by trained healthcare professionals or researchers for retrospective remote monitoring of physiologic parameters in ambulatory individuals 18 years of age and older in home-healthcare environments. As the platform does not provide real-time alerts related to variation of physiologic parameters, users should use professional judgment in assessing patient clinical stability and the appropriateness of using a monitoring platform designed for retrospective review.

The device is intended for continuous data collection supporting intermittent retrospective review of the following physiological parameters:

• Pulse Rate,
• Blood Oxygen Saturation under no-motion conditions,
• Respiratory Rate under no motion conditions,
• Peripheral Skin Temperature,
• Electrodermal Activity,
• Activity associated with movement during sleep

The Empatica Health Monitoring Platform can be used to analyze circadian rhythms and assess activity in any instance where quantifiable analysis of physical motion is desirable.

The Empatica Health Monitoring Platform is not intended for SpO2 monitoring in conditions of motion or low perfusion.

The Empatica Health Monitoring Platform is intended for peripheral skin temperature monitoring, where monitoring temperature at the wrist is clinically indicated.

The Empatica Health Monitoring Platform is not intended for Respiratory Rate monitoring in motion conditions. This device does not detect apnea and should not be used for detecting or monitoring cessation of breathing.

The Empatica Health Monitoring Platform is not intended for Pulse Rate monitoring in patients with chronic cardiac arrhythmias, including atrial fibrillation and atrial/ventricular bigeminy and trigeminy, and is not intended to diagnose or analyze cardiac arrhythmias. The Empatica Health Monitoring Platform is not a substitute for an ECG monitor, and should not be used as the sole basis for clinical decision-making.

The Empatica Health Monitoring Platform is a wearable device and software platform composed by:

• A wearable medical device called EmbracePlus,
• A mobile application running on smartphones called "Care App",
• A cloud-based software platform named "Care Portal".

The EmbracePlus is worn on the user's wrist and continuously collects raw data via specific sensors. These data are wirelessly transmitted via Bluetooth Low Energy to a paired mobile device where the Care App is up and running. The data received are analyzed by one of the Care App software modules, EmpaDSP, which computes the user physiological parameters. Based on the version of the Care App installed, the user can visualize a subset of these physiological parameters. The Care App is also responsible for transmitting, over cellular or WiFi connection sensors' raw data, device information, Care App-specific information, and computed physiological parameters to the Empatica Cloud. On the Empatica Cloud, these data are stored, further analyzed, and accessible by healthcare providers or researchers via a specific cloud-based software called Care Portal.

The Empatica Health Monitoring Platform is intended for retrospective remote monitoring of physiological parameters in ambulatory adults in home-healthcare environments. It is designed to continuously collect data to support intermittent monitoring of the following physiological parameters and digital biomarkers by trained healthcare professionals or researchers: Pulse Rate (PR), Respiratory Rate (RR), blood oxygen saturation (SpO2), peripheral skin temperature (TEMP), and electrodermal activity (EDA). Activity sensors are used to detect sleep periods and to monitor the activity associated with movement during sleep.

The provided FDA 510(k) clearance letter and its attachments describe the acceptance criteria and study that proves the Empatica Health Monitoring Platform (EHMP) meets those criteria, specifically concerning a new Predetermined Change Control Plan (PCCP) for the SpO2 quality indicator (QI) algorithm.

1. Acceptance Criteria and Reported Device Performance

The acceptance criteria are outlined for the proposed modification to the SpO2 Quality Indicator (QI) algorithm. The reported device performance is presented as a statement of equivalence to the predicate device, implying that the acceptance criteria are met, as the 510(k) was cleared.

Note: For the pre-existing functionalities (Pulse Rate, Respiratory Rate, Peripheral Skin Temperature, Electrodermal Activity, Activity and Sleep), the document states that "no changes to the computation... compared with the cleared version" have been introduced, implying their previous acceptance criteria were met and remain valid.

2. Sample Sizes and Data Provenance

• Test Set Sample Size: Not explicitly stated for the SpO2 algorithm modification. The document only mentions "enhancing the development dataset with new samples" for the ML-based algorithm and clinical testing was "conducted in accordance with ISO 80601-2-61... and ... FDA Guidelines for Pulse Oximeters." These standards typically require a certain number of subjects and data points, but the exact numbers are not provided in this public summary.
• Data Provenance: Not specified in the provided document. It does not mention the country of origin, nor whether the data was retrospective or prospective.

3. Number and Qualifications of Experts for Ground Truth

• Number of Experts: Not specified.
• Qualifications of Experts: Not specified. The document states the platform is "intended to be used by trained healthcare professionals or researchers," and later discusses "professional users" and "clinical interpretation," implying that the ground truth for clinical studies would likely involve such experts, but their specific roles, numbers, and qualifications for establishing ground truth are not detailed.

4. Adjudication Method for the Test Set

The adjudication method for establishing ground truth for the test set is not explicitly mentioned in the provided document.

5. Multi Reader Multi Case (MRMC) Comparative Effectiveness Study

There is no mention of a Multi Reader Multi Case (MRMC) comparative effectiveness study being conducted, nor any effect size regarding human readers improving with AI vs. without AI assistance. The device is for "retrospective remote monitoring" by healthcare professionals, implying an AI-driven data collection/analysis with human review, but not necessarily human-AI collaboration in real-time diagnostic interpretation that an MRMC study would evaluate.

6. Standalone (Algorithm Only) Performance

The acceptance criteria for the SpO2 QI algorithm include "Bench testing conducted using a functional tester to simulate a range of representative signal quality issues." This falls under standalone performance, as it tests the algorithm's ability to discriminate data quality without direct human input. Clinical testing also evaluates the algorithm's accuracy (Arms error) in comparison to an established standard, which is also a standalone performance measure.

7. Type of Ground Truth Used

• For the SpO2 QI ML algorithm: The ground truth for low-quality and high-quality data discrimination seems to be an internal standard/reference based on the "FDA-cleared SpO2 algorithm" and potentially expert labeling of data quality during the "enhancing the development dataset."
• For the SpO2 Accuracy (Arms Error): The ground truth for SpO2 values would be established in accordance with ISO 80601-2-61, which typically involves comparing the device's readings against a laboratory co-oximeter or a reference pulse oximeter for arterial oxygen saturation.

8. Sample Size for the Training Set

The document mentions "enhancing the development dataset with new samples" for the ML-based algorithm but does not specify the sample size for the training set.

9. How Ground Truth for Training Set was Established

The ground truth for training the ML-based SpO2 QI algorithm was established by "enhancing the development dataset with new samples." It also mentions performing "feature extraction and engineering on window lengths spanning a 10-30-second range." While it doesn't explicitly state the methodology, given the context of a "binary output" (high/low quality), it implies a labeling process, likely by human experts or based on predefined criteria derived from the previous FDA-cleared algorithm's performance on various data types. For the SpO2 accuracy, the ground truth would typically be established by a reference method consistent with the mentioned ISO standard and FDA guidance.

Ask a specific question about this device

Central Station is a network device, intended to display, record and print monitored physiological data from Nihon Kohden bedside monitors, telemetry receiver and/or transmitters.

Central Station does not perform any data processing on the data received from the Nihon Kohden compatible devices. When Central Station is connected with the Nihon Kohden bedside monitors and telemetry receivers/transmitters the Central Station can:
• Admit and discharge patients on the Nihon Kohden network.
• Display and manage compatible devices' real-time patient clinical data, vital signs, alarms and waveforms.
• Review and trend data calculated by connected Nihon Kohden devices.
• Store and transfer historical clinical data for the connected systems.
• Print patient data.

Central Station is intended for use in professional medical facilities by trained medical personnel.

Central Station is software only product that is installed on a Commercial Off the Shelf (COTS) Computer.

Central Station displays waveforms data and numerical data from a connected bedside monitor, vital sign telemeter, or multiple patient receiver unit on the screen.

Central Station is a network device, intended to provide remote patient monitoring to medical personnel. Central Station displays a list of measured values and a trend graph. Numerical data and various waveforms are color-coded for each parameter. Central Station also has the function of displaying an alarm.

Alarm indication in Central Station is displayed as a result of a judgment by the bedside monitor, vital sign telemeter, or multiple patient receiver units connected to the Central Station. Central Station itself does not have the function to perform alarm indication judgment.

The provided document is a 510(k) clearance letter for the "Central Station" device. This type of document primarily focuses on establishing substantial equivalence to a legally marketed predicate device rather than detailing specific performance acceptance criteria and study results in the same way as a full clinical trial report or a detailed design validation report would.

The document states:

• "Central Station does not perform any data processing on the data received from the Nihon Kohden compatible devices."
• "Alarm indication in Central Station is displayed as a result of a judgment by the bedside monitor, vital sign telemeter, or multiple patient receiver units connected to the Central Station. Central Station itself does not have the function to perform alarm indication judgment."
• "The results of the substantial equivalence assessment, taken together with non-clinical bench testing, software verification, and validation demonstrate that the Central station does not raise concerns regarding its safety and effectiveness compared to its predicate device and operates in accordance with claimed indications for use."

Given these statements, the "Central Station" device is essentially a display, recording, and communication hub. It does not perform diagnostic algorithms or make independent judgments that would typically necessitate the kinds of detailed performance metrics (like sensitivity, specificity, or reader agreement) that are usually established through extensive multi-reader, multi-case (MRMC) studies with expert ground truth. Its primary function is to accurately display and relay data processed by other Nihon Kohden devices.

Therefore, the "acceptance criteria" and "study that proves the device meets the acceptance criteria" for this specific device (Central Station) would primarily revolve around:

• Software Verification and Validation (V&V): Ensuring the software correctly displays, records, and transmits data as designed, and that all features (admit/discharge, trend display, printing, network communication) function as intended without errors.
• Non-Clinical Bench Testing: Confirming interoperability with compatible devices, accuracy of data display, alarm relay, and network functionality.
• Cybersecurity Compliance: Meeting regulatory requirements for cybersecurity.

The document explicitly states that "The software documentation was prepared following the FDA's 'Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices' (released June 14, 2023), specifically for an Enhanced Documentation Level." and "Verification testing was conducted at the system integration level to confirm that the device software fulfills its requirements and that safety and security risk mitigations, where applicable, were effective. Additionally, system-level testing was carried out to show that the software addresses user needs. All unit, integration, and system-level tests successfully met the test protocols."

Based on the provided text, it is not possible to extract the specific quantitative performance metrics (like sensitivity, specificity, or effect sizes for human readers) that would be relevant for a device performing complex data processing or diagnostic functions. The document emphasizes substantial equivalence and basic functional verification, not advanced AI/diagnostic performance validation.

However, I can infer the spirit of typical acceptance criteria and how a device like this would be proven to meet them, guided by the information provided.

Inferred Acceptance Criteria and Device Performance for "Central Station"

Given the device's stated function (display, record, print, and relay data without processing it for diagnosis or alarm judgment), the acceptance criteria would focus on functional correctness, data integrity, interoperability, and system reliability, rather than diagnostic accuracy metrics.

1. Table of Acceptance Criteria and Reported Device Performance

2. Sample Size and Data Provenance

• Test Set Sample Size: Not explicitly stated as a "sample size" in the context of patient data for diagnostic algorithms, because the device doesn't perform diagnostic processing. The "test set" for this device would be a collection of test cases covering all functional requirements, interoperability scenarios with different compatible devices, alarm conditions (relayed), data storage limits, network loads, and user interaction flows. The document mentions "system integration level" and "system-level testing" and that "all unit, integration, and system-level tests successfully met the test protocols." This implies a comprehensive set of non-clinical, bench-level tests.
• Data Provenance: Not applicable in the sense of clinical patient data (e.g., from specific countries, retrospective/prospective studies), as the device does not process primary patient data for diagnosis. The data used for testing would be simulated, generated, or derived from compatible Nihon Kohden monitor outputs in a lab setting to verify the Central Station's display and communication functions.

3. Number of Experts and Qualifications for Ground Truth

• Not Applicable in the traditional sense for diagnostic AI. The ground truth for this device's performance would be the expected output based on its functional specifications and the known inputs from the connected Nihon Kohden devices. For example, if a connected monitor transmits an HR of 70 bpm, the ground truth is that the Central Station must display 70 bpm. These "ground truths" are established by engineering design specifications, not human expert consensus on clinical findings.
• Experts Involved: Software engineers, quality assurance engineers, subject matter experts on the physiological monitoring systems, and potentially clinical users for usability and workflow testing. Their qualifications would be in device design, software development, testing, and clinical application.

4. Adjudication Method for the Test Set

• Not Applicable in the context of clinical interpretation adjudication (e.g., 2+1 radiologist consensus). Adjudication in this context would be internal to the software development and testing process:
  • Test Pass/Fail Criteria: Predetermined pass/fail criteria for each test case.
  • Bug/Defect Resolution: Issues found during testing are logged as bugs, investigated by engineers, and resolved, followed by retesting.
  • Verification Sign-off: Test leads or design engineers review test results and formally sign off on the successful completion of verification.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• No, not performed for this device. An MRMC study is relevant for AI systems that assist human readers in making diagnostic decisions (e.g., radiologists reading X-rays with AI assistance). The Central Station device does not perform any such diagnostic assistance; it merely displays data from other (likely already cleared) devices. Its function is analogous to a monitor or a remote display unit, not a diagnostic AI.

6. Standalone (Algorithm Only) Performance

• Not Applicable. This device does not have a "standalone algorithm" that performs diagnostic or data processing functions independently. Its function is entirely dependent on receiving data from other compatible Nihon Kohden devices.

7. Type of Ground Truth Used

• Functional Specifications and Truth from Connected Devices: The ground truth for this device is based on its functional design specifications (e.g., "display received data," "store data for X hours," "print Y parameters") and the verified output from the connected Nihon Kohden bedside monitors and telemetry systems. It's about data integrity and display accuracy, not clinical outcomes or pathology.

8. Sample Size for the Training Set

• Not Applicable. This device is described as "software only product" that "does not perform any data processing on the data received," and "Central Station itself does not have the function to perform alarm indication judgment." This strongly implies it is a rule-based or deterministic system, not a machine learning/AI system that requires a "training set" in the common sense (i.e., for learning patterns from data for prediction or classification). Therefore, there is no training set.

9. How the Ground Truth for the Training Set Was Established

• Not Applicable. As there is no training set for an AI algorithm, there is no corresponding ground truth establishment process for a training set.

Ask a specific question about this device

The TeleRehab® Aermos Cardiopulmonary Rehabilitation System is intended to acquire and condition the ECG signal from a patient so that it can be transmitted wirelessly from a radiofrequency transmitter to a workstation in a hospital or a clinical setting where the data is displayed and analyzed. This device also measures heart rate and provides visual and audible alarms if the patient's heart rate goes out of a prescribed range. This device is for use with ambulatory adult patients who need monitoring while undergoing cardiovascular and/or pulmonary rehabilitation. The physiological data from monitoring and other patient information (such as patient demographics, exercise protocol and medical information) is stored in a database for tracking and reporting of the patients' progress through rehabilitation.

The TeleRehab® Aermos Cardiopulmonary Rehabilitation System ("Aermos") provides the ECG monitoring functionality required for performing rehabilitation of cardiovascular and/or pulmonary patients. Patients' ECG may be monitored using the Aermos system during exercise under clinical supervision. During monitoring, Aermos provides both visual and audible alarms if the patient's heart rate goes out of a prescribed range. The heart rate alarm indication is one of multiple inputs a clinician may use to modify and adjust rehabilitation activities such as decreasing the patient's level of physical exertion or halting the exercise entirely.

Aermos also provides the ability to plan a patient's rehabilitation program and document the patient's progress through the creation of various types of reports. The report types supported in Aermos include individual treatment plan reports, daily exercise session reports and various patient information reports. Additionally, the Aermos system provides the ability to transfer various report types to the hospital Electronic Medical Records system.

The main components of Aermos are Argus ECG transmitters, the Aermos Workstation and associated networking equipment.

This FDA 510(k) clearance letter pertains to the TeleRehab Aermos Cardiopulmonary Rehabilitation System, which is a device for monitoring ECG signals and heart rate during patient rehabilitation. The provided documentation (the 510(k) Summary) details non-clinical bench testing for performance and safety but explicitly states that clinical testing was not applicable.

Therefore, based on the provided document, the following information regarding acceptance criteria and a study that proves the device meets those criteria, specifically concerning an AI/algorithm-driven component with clinical performance metrics, cannot be fully extracted or is explicitly stated as not performed:

Here's an analysis of the provided information relative to your request:

Acceptance Criteria and Device Performance (Based on Non-Clinical Bench Testing)

Since no clinical study was performed, the "reported device performance" would pertain to the results of non-clinical bench testing against established performance standards. The document does not provide specific quantitative acceptance criteria or reported numerical performance results for the device. Instead, it states that the device's specifications were "verified through internal verification testing" and its usability "evaluated through internal validation testing," and that it complies with various international standards.

Study Details (Based on the provided 510(k) Summary)

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

   • See the table above. Specific quantitative acceptance criteria beyond "compliance with standard" are not provided in this regulatory summary.

2. Sample size used for the test set and the data provenance:

   • The document explicitly states "Clinical Testing: Not applicable."
   • For the non-clinical bench testing, specific sample sizes (e.g., number of devices tested, number of test cases) are not detailed in this 510(k) summary.
   • Data provenance for non-clinical testing would typically be internal laboratory data generated during device development and verification. There is no mention of geographical origin or retrospective/prospective nature as this was not clinical data.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

   • Not applicable, as no clinical study with human interpretation/ground truth was performed. The "ground truth" for bench testing would be defined by validated test equipment and reference standards.

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

   • Not applicable, as there was no study involving human readers or interpretation requiring adjudication.

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 MRMC study was done, as clinical testing was "Not applicable." The device is a physiological signal monitor, not an AI-assisted diagnostic tool that interprets images or signals requiring human reader comparison.

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

   • The core functionality of the device (ECG acquisition, heart rate measurement, alarms) is algorithmic. The performance of these algorithms would have been assessed during the non-clinical bench testing, which is essentially "standalone algorithm" testing against known inputs and expected outputs. Specific quantitative results (e.g., algorithm accuracy for heart rate) are not provided in this summary beyond "compliance with IEC 60601-2-27" and "ANSI/AAMI EC57: 2012, Testing and Reporting Performance Results of Cardiac Rhythm and ST-Segment Measure Algorithms."

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

   • For non-clinical bench testing, the "ground truth" is typically established by:
     • Reference standards and calibrated test equipment: For electrical performance, signal acquisition accuracy, frequency response, etc.
     • Simulated physiological signals: For testing heart rate calculation and alarm thresholds.
     • Design specifications and established engineering principles: For software functionality and cybersecurity.

8. The sample size for the training set:

   • Not applicable. The device is a monitoring system and not primarily driven by a deep learning or machine learning algorithm that requires a "training set" in the sense of a large dataset for model development. The algorithms for heart rate calculation, etc., are likely traditional signal processing algorithms.

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

   • Not applicable, as there was no training set for a machine learning model.

Summary of Device Nature and Regulatory Pathway:

The TeleRehab Aermos Cardiopulmonary Rehabilitation System is a Class II device (Product Codes DRG, DRT) which functions as a physiological signal transmitter and receiver. It monitors ECG and heart rate and provides alarms. Its 510(k) clearance relied on demonstrating substantial equivalence to predicate devices primarily through non-clinical bench testing against recognized industry standards (e.g., IEC 60601 series, ANSI/AAMI, ISO standards) and adherence to FDA guidance documents (e.g., for software, cybersecurity, reprocessing). The explicit statement "Clinical Testing: Not applicable" indicates that the nature of the device and its intended use, combined with the comprehensive non-clinical data, satisfied the FDA's requirements for demonstrating safety and effectiveness without the need for a human-subject clinical study. This is common for devices that are evolutionary improvements on existing technologies with well-understood performance parameters.

Ask a specific question about this device

The Radius VSM and accessories are intended to be used as both a wearable multi-parameter patient monitor and an accessory to a multi-parameter patient monitor that is intended for multi-parameter physiological patient monitoring in hospital and healthcare facilities.

The Radius VSM and accessories are indicated for the monitoring of hemodynamic (including ECG, arrhythmia detection, non-invasive blood pressure, SpO2, Pulse Rate, PVi, heart rate, and temperature), and respiratory (e.g., impedance, acoustic, and pleth-based respiration rate) physiological parameters along with the orientation and activity of adults.

The Radius VSM and accessories are indicated for the non-invasive continuous monitoring of functional oxygen saturation of arterial hemoglobin (SpO2) and Pulse Rate (PR) of well or poorly perfused adults during both no motion and motion conditions.

The Radius VSM and accessories are indicated for continuous monitoring of skin temperature of adults.

The Radius VSM and accessories are indicated for monitoring of the orientation and activity of patients including those susceptible to pressure ulcers.

The Radius VSM and accessories are indicated for the continuous non-invasive monitoring of PVi as a measure of relative variability of the photoplethysmograph (pleth) of adults during no motion conditions. PVi may be used as a noninvasive dynamic indicator of fluid responsiveness in select populations of mechanically ventilated adult patients. Accuracy of PVi in predicting fluid responsiveness is variable and influenced by numerous patient, procedure and device related factors. PVi measures the variation in the plethysmography amplitude but does not provide measurements of stroke volume or cardiac output. Fluid management decisions should be based on a complete assessment of the patient's condition and should not be based solely on PVi.

Devices with Masimo technology are only indicated for use with Masimo accessories.

Radius VSM Accessories:

Radius VSM ECG Electrodes are disposable, single-patient use ECG electrodes intended to acquire ECG signals from the surface of the body. They are indicated for use on adults for up to 3 days of skin surface contact.

Radius VSM Blood Pressure Cuffs are accessories intended to be used with a noninvasive blood pressure measurement system to measure blood pressure. They are indicated for use on adults during no motion conditions.

The Radius VSM and accessories are an FDA cleared (K223498), wearable, battery-operated, multi-modular patient monitoring platform that allows for the ability to scale and tailor the use of different monitoring technologies based upon the hospital and clinician's assessment of what technologies are appropriate.

As part of this submission, a MAP feature is being added to the Radius VSM. The feature is a software feature that uses the previously cleared systolic and diastolic measurement capabilities to automate the calculation of MAP using the following formula: MAP = 1/3* Systolic + 2/3*Diastolic.

The MAP is calculated by the Radius VSM NIBP Module and displayed on the Radius VSM Wearable Monitor. There were no other features added as part of this submission.

The provided 510(k) clearance letter and summary discuss the addition of a Mean Arterial Pressure (MAP) feature to the previously cleared Radius VSM and Accessories device. The primary focus of the performance data section is on validating this new MAP feature.

Here's an analysis of the acceptance criteria and the study proving the device meets them, based on the provided document:

Acceptance Criteria and Reported Device Performance

The document states that the acceptance criterion for Blood Pressure (including MAP) is:

"Meets ISO 81060-2 (Mean difference of ≤5 mmHg with a standard deviation of ≤8 mmHg)"

The document directly states that the results of the clinical testing supported the clinical performance of the MAP in accordance with ISO 81060-2. While specific numerical results (e.g., the exact mean difference and standard deviation achieved) are not explicitly provided in the summary table, the clearance implies that these metrics fell within the specified ISO 81060-2 limits for the MAP feature.

Table 1: Acceptance Criteria and Reported Device Performance for MAP Feature (as inferred from the document)

Study Details for MAP Feature Validation

1. Sample Size Used for the Test Set and Data Provenance:

   • Sample Size: The document does not explicitly state the numerical sample size (number of subjects/patients) used for the clinical test set. It only mentions "clinical study data."
   • Data Provenance: The document does not specify the country of origin. It indicates it was a "clinical study" and implies it was prospective ("clinical testing is provided to support its performance" for the added feature).

2. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications of those Experts:

   • Not applicable as the ground truth was established by an objective reference device, not human experts.

3. Adjudication Method for the Test Set:

   • Not applicable, as the method for ground truth establishment was comparison to a reference device.

4. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study was done:

   • No, an MRMC study was not done. The study was a comparison of the device's calculated MAP to invasively measured MAP from a reference device. This is a technical performance validation, not a study assessing human reader improvement with AI assistance.

5. If a Standalone Performance (i.e., algorithm only without human-in-the-loop performance) was done:

   • Yes, this was a standalone performance study. The Radius VSM automatically calculates the MAP based on the NIBP measurements (Systolic and Diastolic Pressure). The clinical testing validated the accuracy of this calculation against a reference standard, without human intervention in the MAP calculation or interpretation for the test itself.

6. The Type of Ground Truth Used:

   • Reference Ground Truth: Invasively measured MAP values from a 510(k) cleared reference device (K171801). This reference device is identified as "IntelliVue Multi-Measurement Module X3." This constitutes a device-based reference standard or instrument-based ground truth.

7. The Sample Size for the Training Set:

   • The document does not provide information about a training set since the MAP feature appears to be a direct calculation using a standard formula (MAP = 1/3* Systolic + 2/3*Diastolic) rather than a machine learning model that requires a training phase. While the device as a whole (Radius VSM) likely had training and validation phases for its other parameters, the specific "addition of a Mean Arterial Pressure (MAP) feature" is described as a software feature that "automates the calculation" using a known formula. Therefore, a separate training set for this specific MAP feature is unlikely to have been required or used in the conventional machine learning sense.

8. How the Ground Truth for the Training Set was Established:

   • As inferred above, a specific training set and ground truth establishment for this isolated MAP calculation feature are not described, given its nature as a direct formulaic calculation.

Summary of Key Information:

The core of this submission revolves around adding a simple, formula-based calculation for MAP. The primary study presented is a clinical validation confirming that the device's computed MAP aligns with a known industry standard (ISO 81060-2) when compared against an invasive reference device. This is a technical performance validation rather than a complex AI-driven diagnostic study.

Ask a specific question about this device