Search Results

The IMRACT Pager System is intended to be used for the annunciation of events within a zone coverage area that is secondary to the primary care that is provided. Primary care via the central station, patient bedside monitor, or telemetry system remains unchanged. This device is intended to be used within the hospital/facility environment.

The IMPACT pager is intended for the annunciation of events to provide zone information that is secondary to the primary care provided. It provides specific information within a zone and alerts or draws an identified individual's aftention to a defined patient condition in a timely manner. Primary care via the central station, telemetry system, or patient bedside monitor remains unchanged by the addition of the IMPACT pager.

The provided 510(k) summary for the Marquette IMPACT Pager System focuses on establishing substantial equivalence to a predicate device rather than presenting a performance study with detailed acceptance criteria and standalone performance for a medical device that makes diagnostic or treatment decisions.

The IMPACT Pager System is described as a device for "annunciation of events to provide zone information that is secondary to the primary care provided." It is explicitly stated that "Primary care via the central station, telemetry system, or patient bedside monitor remains unchanged by the addition of the IMPACT pager." This indicates that the device does not perform primary diagnostic or therapeutic functions, but rather relays information.

Therefore, the typical metrics like sensitivity, specificity, accuracy, or MRMC study results, which are common for AI/ML-driven diagnostic or prognostic devices, are not applicable or reported for this type of device. The "performance" here refers more to reliable functionality and information transmission.

Here's the breakdown of the information that is available based on the provided text, and where some requested information is not applicable (N/A) or not provided:

Acceptance Criteria and Study for Marquette IMPACT Pager System (K971868)

The submission indicates that the device is a "pager system" for "annunciation of events" and is "secondary to the primary care provided." The performance testing focused on functional correctness and safety, rather than diagnostic accuracy.

1. Table of Acceptance Criteria and Reported Device Performance

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

• Sample Size for Test Set: Not explicitly specified. The text mentions "functional performance testing" and "validation test results" indicating a test set was used, but the number of test cases, durations, or specific scenarios are not provided in the summary.
• Data Provenance: Not specified, but likely internal testing performed by Marquette Medical Systems. As it's a pager system, the "data" would be the simulated or live alarm/event information being transmitted.
• Retrospective/Prospective: Neither term is used. The testing described appears to be prospective functional and safety testing conducted by the manufacturer.

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

• Number of Experts: Not applicable or not specified. Given the nature of the device (a pager for event annunciation), "ground truth" would be the correct transmission of information. The "correctness" was likely verified against the source (primary monitoring system) by technical personnel, not by medical experts establishing a diagnostic ground truth.
• Qualifications of Experts: N/A for diagnostic "ground truth." Marquette's "Technical Coordinator" reviewed UL listings, and "reliability group" performed safety tests.

4. Adjudication Method for the Test Set

• Adjudication Method: Not applicable or not explicitly described. Since the "truth" is the direct replication of information from a primary system, adjudication by multiple human readers for diagnostic accuracy is not relevant to this device's function.

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

• MRMC Study: No, an MRMC study was not done. This type of study is for evaluating the impact of AI on human reader performance for diagnostic tasks, which is not the function of a pager system.
• Effect Size of Human Readers Improve with AI vs. without AI assistance: Not applicable.

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

• Standalone Performance: While the "CardioPager replicates arrhythmia type, heart rate, and waveform information..." and "this information was compared for correctness during the functional performance testing," this isn't a "standalone" performance in the AI/ML sense (e.g., reporting sensitivity/specificity of an algorithm making a diagnosis). It's a functional test of the system's ability to accurately transmit information. No specific metrics like sensitivity or specificity for its own "interpretations" (as it doesn't interpret, it replicates) are provided. The "algorithm" here is essentially the software logic for data transmission and display.

7. The Type of Ground Truth Used

• Type of Ground Truth: The ground truth for the functional performance testing was the output of the primary patient monitoring system (e.g., "arrhythmia type, heart rate, and waveform information obtained from the monitoring system"). The pager's output was compared against this primary source for correctness.

8. The Sample Size for the Training Set

• Sample Size for Training Set: Not applicable / Not specified. This device does not appear to be an AI/ML device that requires a training set in the conventional sense. Its function is information relay, not learning from data to perform a diagnostic or predictive task.

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

• How Ground Truth for Training Set Was Established: Not applicable. There is no mention of a training set or an AI/ML component that requires training data in the provided summary.

Ask a specific question about this device

The Marquette Eagle 4000 Patient Monitor is a multi-parameter patient monitoring system that is cleared to market as part of Marquette's wired Unity network. The Marquette Eagle 4000 Patient Monitor is now indicated for use on Marquette's wireless LAN.

Wireless LAN is a method of communicating to the Marquette Unity Network using a wireless connection. This capability:

~provides identical networking function to standard Ethernet;

~permits integration to the network without physical connection to the Ethernet wall plate.

This device is viewed as a technology change and can be used with various other Marquette devices. Marquette has clearances for various patient monitors and its central station which are part of its wired Unity network. The intended patient population is limited to the same population as the device that the technology is being used with.

This device should be used by people who are trained in the use of the equipment.

This device is intended to be used within the hospital / facility environment.

The Marquette Eagle 4000 Patient Monitor is a multi-parameter patient monitoring system that is cleared to market as part of Marquette's wired Unity network. The Marquette Eagle 4000 Patient Monitor is now indicated for use on Marquette's wireless LAN.

Wireless LAN is a method of communicating to the Marquette Unity Network using a wireless connection. This capability:

~provides identical networking function to standard Ethernet; ~permits integration to the network without physical connection to the Ethernet wall plate.

The provided text describes a 510(k) submission (K971683) for a Wireless LAN (Local Area Network) - wireless Ethernet technology change, intended for use with Marquette Medical Systems' Eagle 4000 Patient Monitor. This document is focused on demonstrating the safety and effectiveness of the wireless networking capability, not on the performance of a diagnostic AI device.

Therefore, many of the requested elements for describing an AI device's performance acceptance criteria and study (such as sample sizes for training/test sets, ground truth establishment, expert qualifications, MRMC studies, standalone performance, and effect size with AI assistance) are not applicable to this submission.

The primary focus of this submission is to demonstrate that the wireless LAN, when integrated with an already cleared patient monitor, maintains the same level of safety and effectiveness as the wired version.

Here's an attempt to answer the applicable points based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

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

The document indicates "Verification and validation testing was done on the wireless LAN in use with the Eagle 4000 Patient Monitor." However, it does not specify a sample size or data provenance in terms of patient data or case numbers. The testing appears to be focused on device-centric performance parameters (safety, EMC, and functional equivalence) rather than clinical patient data. The provenance for the testing is Marquette Medical Systems, based in Milwaukee, WI, USA. The testing context is implicitly prospective as it's about verifying a new technology integration.

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

Not applicable. The "ground truth" for this submission revolves around engineering and regulatory standards for device safety, EMC, and networking functionality. It does not involve establishing a clinical "ground truth" requiring medical experts for diagnosis or outcome measurement. The "truth" is established by compliance to international and national standards by qualified engineers and testers.

4. Adjudication method for the test set

Not applicable. There is no indication of clinical adjudication or expert consensus methods for this type of technical performance testing.

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 submission is for a wireless networking component of a patient monitor, not an AI diagnostic device. No human-in-the-loop studies or AI-driven performance enhancements are discussed.

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

Not applicable. This describes a hardware and software networking integration, not a standalone algorithm. The device's "performance" is its ability to transmit data reliably and safely, equivalent to a wired connection, and to comply with relevant safety and EMC standards.

7. The type of ground truth used

The "ground truth" for this submission refers to the compliance with established international and national technical standards (IEC 601-1-1, IEEE C95.1-1991, FCC certification, CISPR 11 Class A, IEC 1000-4 series, MIL-STD 462D) and the functional equivalence of networking capabilities between wired and wireless configurations.

8. The sample size for the training set

Not applicable. This device integrates a wireless LAN, it is not an AI algorithm that requires a training set of data.

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 ground truth established in this context.

Ask a specific question about this device

PHi-Res is a software option for Marquette ECG analysis systems for high resolution P wave analysis. PHi-Res is intended to be used in a hospital or clinic environment by health care professionals for recording low amplitude/high frequency components of the surface electrocardiogram for P wave analysis.

PHi-Res analysis is only intended to provide the measurements of the signal averaged P wave and is not intended to provide any interpretation of those measurements or any kind of diagnosis. The P wave measurements provided by PHi-Res analysis are intended to be used by qualified personnel in evaluating the patient in conjunction with patient's clinical history, symptoms, other diagnostic tests, as well as the professional's clinical judgment.

PHi-Res is intended for patient populations including adult and pediatric.

PHi-Res analysis is a software option for Marquette MAC-series electrocardiographs for high resolution P wave analysis.

The provided document, K972199, details a 510(k) submission for the Signal-Averaged High Resolution P Wave Analysis (PHi-Res) Option. However, it does not contain specific acceptance criteria or an explicit study proving the device meets said criteria in a quantifiable manner as per the requested information. Instead, it relies on general quality assurance measures and substantial equivalence to a predicate device.

Here's an analysis of the information that is available, and what is missing based on your request:

1. Table of Acceptance Criteria and Reported Device Performance

Missing Information: The document states that PHi-Res is "as safe, as effective, and performs as well as the predicate device," but it does not provide quantifiable acceptance criteria (e.g., minimum sensitivity, specificity, or error rates) or specific data demonstrating this performance for PHi-Res itself. It relies on the assertion of substantial equivalence.

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

• Sample Size: Not specified.
• Data Provenance: Not specified (e.g., country of origin, retrospective/prospective).
  The document mentions "software testing and field tests," but no details on the size or nature of these tests are provided.

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

• Number of Experts: Not specified.
• Qualifications of Experts: Not specified.
  The document refers to the data being used by "qualified personnel" and "competent health professionals," but this relates to the intended use of the device, not the establishment of ground truth for testing.

4. Adjudication method for the test set

• Adjudication Method: Not specified.

5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance

• MRMC Study: No, this type of study was not reported.
  The device provides P-wave measurements and is not intended to provide interpretation or diagnosis. Therefore, a study on human reader improvement with AI assistance would not be applicable to this device's stated function.

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

• Standalone Performance: Implied, but no specific study or metrics are provided. The device's primary function is to provide automated P-wave measurements. The "software testing" and "field tests" would likely assess this standalone performance against some internal standard or the predicate device, but no details are given.

7. The type of ground truth used

• Type of Ground Truth: Not specified.
  Given the reliance on substantial equivalence to the predicate device and the "software testing and field tests," the ground truth likely involved comparing PHi-Res measurements to those from the predicate device or a gold standard ECG measurement technique, but this is not explicitly stated.

8. The sample size for the training set

• Training Set Sample Size: Not applicable/not specified.
  The document states the PHi-Res analysis "employs the same technology as the predicate device." There is no indication that this device uses machine learning or AI that would require a separate "training set" in the modern sense. It appears to be a software option based on established signal processing algorithms.

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

• Ground Truth for Training Set: Not applicable/not specified, as there is no indication of a training set for an AI/ML model.

Summary of Device and Evidence:

The PHi-Res device is a software option for Marquette MAC-series electrocardiographs designed for high-resolution P-wave analysis. Its purpose is to perform signal averaging of the atrial wave and provide its measurements; it explicitly states it is not intended to provide interpretation or diagnosis.

The "study" or evidence provided to support its safety and effectiveness is based on the following:

• Quality Assurance Measures: Requirements specification reviews, code inspections, software testing, and field tests. The results of these measures "demonstrated that PHi-Res analysis is as safe, as effective, and performs as well as the predicate device, Marquette High Resolution Option for MAC-Series Electrocardiographs."
• Substantial Equivalence: The submission heavily relies on demonstrating substantial equivalence to a predicate device (Marquette High Resolution ECG Option for MAC-Series Electrocardiographs) by stating it "employs the same technology" and performs "as well as" it.

In conclusion, the document provided is a 510(k) summary focused on demonstrating substantial equivalence through general quality assurance processes rather than a detailed performance study with specific acceptance criteria, sample sizes, and ground truth methodologies that are common in more recent medical device submissions, especially those involving AI/ML.

Ask a specific question about this device

CardioServ is intended to be used for the emergency resuscitation of cardiac arrest victims and clinical cardiac dysrhythmia.
CardioServ is intended to be used by trained operators
CardioServ is designed for external and internal defibrillation (including cardioversion)
CardioServ is capable of monitoring the heart rate with adjustable alarm limits.
CardioServ is designed for external pacing with adjustable current and frequency.
The intended use of CardioServ is identical to the intended use of the predicate devices.

CardioServ is a portable defibrillator with ECG monitor, built-in recorder, and the capability of external pacing with adjustable current and frequency.

The provided text describes a medical device, the CardioServ P, and its equivalence to predicate devices, but it does not contain the specific information requested about acceptance criteria and a study proving those criteria are met for AI/algorithm performance.

The document is a 510(k) summary from 1997, focusing on the safety and effectiveness of a defibrillator/pacemaker. The "Technology" section mentions compliance with voluntary standards (e.g., ANSI/AAMI, IEC) and quality assurance measures like "software and hardware testing," "safety testing," and "final validation testing by an independent test group." However, these are general development and testing processes for a hardware medical device, not a study evaluating an AI algorithm's performance against specific acceptance criteria.

Therefore, I cannot provide the requested information. The document focuses on the equivalence of the CardioServ P to predicate devices based on its intended use, technological similarity, and adherence to established electrical and medical device standards. There is no mention of an AI component, performance metrics like sensitivity/specificity, or a study design suitable for evaluating such a component.

To fulfill your request, I would need a document detailing the performance evaluation of an AI or algorithm, including specific acceptance criteria and the results of a study designed to demonstrate compliance.

Ask a specific question about this device

Responder 2500 is intended to be used in a hospital environment by emergency personnel in evaluating and treating life-threatening arrhythmias.
♦ Responder 2500 is intended to be used by trained operators
♦ Responder 2500 is designed for external and internal defibrillation (including cardioversion)
♦ Responder 2500 is capable of monitoring the heart rate with adjustable alarm limits.
♦ Responder 2500 is designed for external pacing with adjustable current and rate.

Responder 2500 and Cardiac Care System is an emergency cardiac monitor/defibrillator with an external pacing option.

This 510(k) submission for the Marquette Electronics Responder 2500 Defibrillator and Cardiac Care System focuses on demonstrating substantial equivalence to predicate devices (Marquette Series 900 and Marquette Series 1500 Responder) rather than providing detailed acceptance criteria and a study proving device performance against those criteria in a typical AI/ML medical device context.

Therefore, many of the requested details regarding acceptance criteria, sample sizes, expert involvement, and ground truth establishment, which are standard for AI/ML device evaluations, are not present in this document. The submission relies on compliance with voluntary standards and general performance equivalency to predicate devices.

Here's an attempt to extract and interpret the available information according to your requested format, along with explicit statements about what information is missing:

1. Table of Acceptance Criteria and Reported Device Performance

Study Description:

The document describes the "development" of the Responder 2500, which involved various testing and quality assurance measures to confirm compliance with listed voluntary standards and establish equivalence to predicate devices. These measures are:

• Requirements specification reviews
• Code inspections
• Software and hardware testing
• Safety testing
• Environmental testing
• Field tests on the monitor function
• Animal studies on the defibrillation and pacing functions
• Final validation

The results of these measurements "demonstrated that Responder 2500 is as safe, as effective, and performs as well as the predicate devices Marquette Series 900 and Marquette Series 1500 Responder."

Missing Information (Relative to AI/ML Device Evaluation):

Specific quantitative acceptance criteria (e.g., sensitivity, specificity, AUC thresholds) are not provided, as this is a traditional medical device submission focused on performance standards and predicate equivalence.

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

• Sample Size: Not specified. The document mentions "field tests on the monitor function" and "animal studies on the defibrillation and pacing functions," but provides no numbers for test subjects (human or animal) or the duration/scope of these tests.
• Data Provenance: Not specified. No mention of country of origin or whether data was retrospective or prospective.

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)

• Not applicable/Not specified. The evaluation relies on compliance with established engineering standards and functional performance, not on expert adjudication of diagnostic outcomes like an AI/ML algorithm might.

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

• Not applicable/Not specified. (See point 3)

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. This is not an AI/ML device and therefore no MRMC study, or impact on human readers, was conducted or reported.

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

• The device itself is a standalone functional unit. Performance was assessed based on its functional capabilities (monitoring, defibrillation, pacing) and compliance with standards. There isn't an "algorithm only" performance concept in the way it applies to AI/ML devices, as the device's function is inherently a standalone "performance."

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

• Functional Ground Truth: The "ground truth" for this device's performance would be defined by the technical specifications outlined in the voluntary standards referenced (e.g., ANSI/AAMI DF2-1996 for defibrillator performance, IEC 601-2-27 for ECG performance). For animal studies, the "ground truth" would be established by physiological measurements and successful or unsuccessful defibrillation/pacing events as determined by veterinarians or other qualified personnel. It is not expert consensus for image interpretation or pathology.

8. The sample size for the training set

• Not applicable. This device is a hardware/software system, not a machine learning model that undergoes a "training" phase with a dataset in the AI/ML sense. Its development involved traditional engineering, software development, and quality assurance processes.

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

• Not applicable. (See point 8)

Ask a specific question about this device

The pager is intended for the annunciation of events to provide zone information that is secondary to the primary care provided. It provides specific information within a zone and alerts or draws an identified individual's attention to a defined patient condition in a timely manner. Primary care via the central station, telemetry system, or patient bedside monitor remains unchanged by the addition of a pager.

This device is commonly known as a pager and an alarm display unit.

This document is a 510(k) summary for a pager and alarm display unit, submitted in 1996. It explicitly states that performance standards have not yet been established for this device.

Therefore, based on the provided text, it's not possible to define specific acceptance criteria, provide a table of performance, or describe a study that proves the device meets such criteria because:

• No specific acceptance criteria are listed. The document states that the device is "unclassified" and that "performance standards have not yet been established."
• No detailed study is described. The document merely states "Validation test results indicate that the pager meets the requirements of its intended use" without providing any details about the methodology, sample size, or specific metrics used.

In a modern context for a medical device with an AI component, the requested information would be crucial. However, for this 1996 pager submission, the regulatory requirements and the nature of the device (a simple communication tool) were significantly different.

Therefore, I cannot provide the requested information for this specific device as it is not present in the provided 510(k) summary.

Ask a specific question about this device

The Marquette Eagle 4000 Patient Monitor is designed to monitor and display patient data. Its design allows the operator to adjust parameter alarm settings that would audibly and visually notify the operator when a violation occurs. The option is provided for printing of information by a paper recorder. Use of the Marquette Eagle 4000 Patient Monitor is intended for patient populations including: adult, pediatric, and/ or neonatal. Use of the Marquette Eagle 4000 Patient Monitor is not recommended for use in patient's home or residence, during patient transport, or when it has not been ordered by a physician or other qualified medical personnel. Use of the Marquette Eagle 4000 Patient Monitor is intended for operating room (OR), post anesthesia recovery, critical care, surgical intensive care, respiratory intensive care, coronary care, medical intensive care, pediatric intensive care, or neonatal intensive care. These departments are typically located in hospitals or may be located in outpatient clinics or free standing surgical centers. It is intended for use by physicians, physician assistants, registered nurses, certified registered nurse anesthetists, or other hospital personnel trained in the use of the equipment.

The Marquette Eagle 4000 Patient Monitor is a patient monitoring system that is designed to be used to monitor a patient's basic physiological parameters including: electrocardiography (ECG), invasive blood pressure, non-invasive blood pressure, oxygen saturation, temperature, respiration, apnea detection, pulse rate, cardiac output, and full arrhythmia analysis. The device now includes Marquette's 12 lead ECG Analysis program (commonly referred to as 12 SL).

The provided text is a 510(k) summary for the Marquette Eagle 4000 Patient Monitor, dated February 2, 1997. This document focuses on demonstrating substantial equivalence to predicate devices for regulatory clearance, rather than conducting a detailed performance study with specific acceptance criteria and detailed quantitative results as might be found in a modern clinical trial or a more recent AI/ML device submission.

Therefore, much of the requested information cannot be extracted directly from this document because it predates the rigorous, quantitative evaluation standards common for AI/ML medical devices today. The summary primarily relies on a general statement of "verification and validation testing" and "accuracy requirements as specified in the contents of the premarket notification submission" without detailing the specifics of these tests or their results.

Here's a breakdown of what can and cannot be answered based on the provided text:

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

• Acceptance Criteria: Not explicitly stated in a quantifiable way in this summary. The summary broadly states that "Test results indicate that the Eagle 4000 Patient Monitor provides an equivalent level or better in performance, when compared to the legally marketed predicate devices when tested to the accuracy requirements as specified in the contents of the premarket notification submission." The "accuracy requirements" themselves are not provided.
• Reported Device Performance: No specific quantitative performance metrics (e.g., sensitivity, specificity, accuracy, error rates) are reported in this summary.

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

• Sample Size: Not specified.
• Data Provenance: Not specified.

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)

• Not applicable/Not specified. The document does not describe the establishment of ground truth by independent experts in a test set, as would be common for AI/ML devices. The "verification and validation testing" likely refers to internal engineering and clinical validation testing against established standards or predicate device performance.

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

• Not applicable/Not specified. There is no mention of an adjudication process for a test set.

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. This type of study (MRMC for AI assistance) is not mentioned and is highly unlikely given the device type (patient monitor) and the era (1997). The device provides physiological parameter monitoring and 12-lead ECG analysis, but there's no indication of it being an AI-assisted diagnostic tool that would improve human reader performance in the modern sense.

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

• Yes, implicitly. The performance evaluation would inherently be of the device's algorithms and hardware alone, as it's a patient monitor providing data and analysis. The "verification and validation testing" would assess the accuracy of its measurements and analyses (e.g., ECG, blood pressure, arrhythmia detection) as a standalone system. However, specific results are not provided.

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

• Not explicitly defined in the summary. For a patient monitor, ground truth would typically involve:
  • Reference Devices: Comparing measurements against highly accurate, calibrated reference instrumentation (e.g., for blood pressure, temperature, oxygen saturation).
  • Validated ECG Databases: For the "12 SL" ECG analysis program, ground truth might involve comparisons against widely accepted ECG databases with expert-adjudicated diagnoses, though this is not stated.
  • Clinical Observation: For arrhythmia detection, comparison against expert interpretation of concurrent ECG recordings.
• The summary only broadly refers to "accuracy requirements as specified in the contents of the premarket notification submission."

8. The sample size for the training set

• Not applicable/Not specified. The document does not mention "training sets" as it would for a machine learning device. The ECG analysis program would have been developed using a dataset, but it's not characterized as a "training set" in the context of modern ML.

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

• Not applicable/Not specified. As no "training set" is described, its ground truth establishment is not discussed.

In summary, the provided 510(k) summary from 1997 is a regulatory declaration of substantial equivalence for a patient monitor and does not contain the detailed performance study information, acceptance criteria, or ground truth methodologies that are now standard for AI/ML device submissions. The document's purpose is to demonstrate that the Eagle 4000 is as safe and effective as its predicate devices based on general verification and validation testing, without providing the quantitative specifics you've requested.

Ask a specific question about this device

CardioServ is intended to be used for the emergency resuscitation of cardiac arrest victims and clinical cardiac dysrhythmia.
♦ CardioServ is intended to be used by trained operators
♦ CardioServ is designed for external and internal defibrillation (including cardioversion)
♦ CardioServ is capable of monitoring the heart rate with adjustable alarm limits.
♦ CardioServ is capable of monitoring the saturation of O₂ with adjustable alarm limits.
The intended use of CardioServ is identical to the intended use of the predicate devices.

CardioServ S is a portable defibrillator with ECG monitor, built-in recorder, and the capability of monitoring the saturation of O₂ with adjustable alarm limits.

The provided text describes a medical device, the "CardioServ S," which is a portable defibrillator with ECG monitor and O₂ saturation monitoring capabilities. However, the document does not contain any information regarding specific acceptance criteria, a study proving the device meets those criteria, or details related to an AI/machine learning component.

The "Performance" section explicitly states that "CardioServ complies with the voluntary standards ANSI/AAMI DF2-1989, ANSI/AAMI ES1-1993. IEC 606-1. IEC 601-1-1, IEC 601-2-4, and prEN865." It also mentions "Requirements specification reviews, code inspections, software and hardware testing, safety testing, environmental testing, final validation testing by an independent test group, field tests" as quality assurance measures. The conclusion is that "CardioServ is as safe, as effective, and performs as well as the predicate devices CardioServ SCP 910 and NELLCOR Escort 300 Series."

Therefore, based on the provided text, I cannot fulfill your request for the following information:

• A table of acceptance criteria and the reported device performance
• Sample size used for the test set and the data provenance
• Number of experts used to establish the ground truth for the test set and their qualifications
• Adjudication method
• If a multi-reader multi-case (MRMC) comparative effectiveness study was done
• If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
• The type of ground truth used
• The sample size for the training set
• How the ground truth for the training set was established

The document focuses on regulatory compliance through adherence to voluntary standards and comparison to predicate devices, rather than a performance study with specific metrics and acceptance criteria as described in your request, which are typically associated with AI/ML-driven medical devices.

Ask a specific question about this device

The Marquette Eagle 3000 Patient Monitor is designed to monitor and display patient data. Its design allows the operator to adjust parameter alarm settings that would audibly and visually notify the operator when a violation occurs. The option is provided for the printing of information by a paper recorder. Use of the Marquette Eagle 3000 Patient Monitor is intended for patient populations including: adult, pediatric, and/ or neonatal. Use of the Marquette Eagle 3000 Patient Monitor is not recommended for use in patient's home or residence, during patient transport, or when it has not been ordered by a physician or other qualified medical personnel. Use of the Marquette Eagle 3000 Patient Monitor is intended for operating room (OR), post anesthesia recovery, critical care, surgical intensive care, respiratory intensive care, coronary care, medical intensive care, pediatric intensive care, or neonatal intensive care. These departments are typically located in hospitals or may be located in outpatient clinics or free standing surgical centers. It is intended for use by physicians, physician assistants, registered nurses, certified registered nurse anesthetists, or other hospital personnel trained in the use of the equipment.

The Marquette Eagle 3000 Patient Monitor is a patient monitoring system that is designed to be used to monitor a patient's basic physiological parameters including: electrocardiography (ECG), invasive blood pressure, non-invasive blood pressure, temperature, and arrhythmia analysis. The Eagle 3000 Patient Monitor is a factory configured or fixed-parameter patient monitoring device. Options which allow for the monitoring of invasive blood pressure and the printing of information by a paper recorder may be added to the basic monitor configuration at the factory prior to shipment to the customer. The Eagle 3000 Patient Monitor has NOT been designed as a modular patient monitoring device which allows the addition and deletion of monitoring parameters at the point-of-use. The Eagle 3000 Patient Monitor has NOT been designed as a transport monitor. Therefore, no battery power feature or option has been included with this device.

The provided text does not contain detailed acceptance criteria or a study proving the device meets those criteria in the format requested. The document is a 510(k) summary, which focuses on demonstrating substantial equivalence to a legally marketed predicate device rather than presenting a detailed performance study with specific acceptance criteria and their fulfillment.

Here's a breakdown of what can be extracted and what is missing:

The document states:

• "Comparative testing was done on the Eagle 3000 Patient Monitor and its predicate devices. Precision and accuracy testing was also performed. Test results indicate that the Eagle 3000 Patient Monitor provides an equivalent level or better in performance, when compared to the legally marketed predicate devices when tested to the accuracy requirements as specified in the contents of the premarket notification submission."

This indicates that some form of testing was performed, but the specifics requested in your prompt (like actual acceptance criteria values, reported performance, sample sizes, ground truth details, etc.) are not included in this summary.

Therefore, I cannot populate the table or answer most of the questions based on the provided text.

However, I can extract the following limited information:

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

   • Acceptance Criteria: Not explicitly stated with numerical values. The summary only mentions "accuracy requirements as specified in the contents of the premarket notification submission."
   • Reported Device Performance: "Equivalent level or better in performance, when compared to the legally marketed predicate devices." No specific numerical performance metrics are provided.

2. Sample size used for the test set and the data provenance: Not available in the provided text.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not available in the provided text.

4. Adjudication method (e.g., 2+1, 3+1, none) for the test set: Not available in the provided text.

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 diagnostic algorithm, and the summary does not describe an MRMC study or AI assistance.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: Not explicitly stated, as the nature of the "testing" is only broadly described as "comparative testing" and "precision and accuracy testing." Given it's a monitor, its performance is inherently "standalone" in its measurement capabilities, but specific details are missing.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.): Not available in the provided text.

8. The sample size for the training set: Not applicable. This is a 1996 device and summary, and it's a patient monitor, not an AI/ML algorithm that requires a "training set" in the modern sense.

9. How the ground truth for the training set was established: Not applicable for the same reasons as point 8.

Ask a specific question about this device

The Marquette Eagle 3000 Patient Monitor is designed to monitor and display patient data. Its design allows the operator to adjust parameter alarm settings that would audibly and visually notify the operator when a violation occurs. The option is provided for the printing of information by a paper recorder. Use of the Marquette Eagle 3000 Patient Monitor is intended for patient populations including: adult, pediatric, and/ or neonatal. Use of the Marquette Eagle 3000 Patient Monitor is not recommended for use in patient's home or residence, during patient transport, or when it has not been ordered by a physician or other qualified medical personnel. Use of the Marquette Eagle 3000 Patient Monitor is intended for operating room (OR), post anesthesia recovery, critical care, surgical intensive care, respiratory intensive care, coronary care, medical intensive care, pediatric intensive care, or neonatal intensive care. These departments are typically located in hospitals or may be located in outpatient clinics or free standing surgical centers. It is intended for use by physicians, physician assistants, registered nurses, certified registered nurse anesthetists, or other hospital personnel trained in the use of the equipment.

The Marquette Eagle 3000 Patient Monitor is a patient monitoring system that is designed to be used to monitor a patient's basic physiological parameters including: electrocardiography (ECG), invasive blood pressure, non-invasive blood pressure, oxygen saturation, temperature, and end tidal carbon dioxide (EtCO2). PA Wedge is also available. The Eagle 3000 Patient Monitor is a factory configured or fixed-parameter patient monitoring device. Options which allow for the monitoring of invasive blood pressure and the printing of information by a paper recorder may be added to the basic monitor configuration at the factory prior to shipment to the customer. The Eagle 3000 Patient Monitor has NOT been designed as a modular patient monitoring device which allows the addition and deletion of monitoring parameters at the point-of-use. The Eagle 3000 Patient Monitor has NOT been designed as a transport monitor. Therefore, no battery power feature or option has been included with this device.

Marquette Electronics, Inc. did not provide quantifiable acceptance criteria or device performance in the provided 510(k) summary (K960272). The summary broadly states that "Comparative testing was done on the Eagle 3000 Patient Monitor and its predicate devices. Precision and accuracy testing was also performed. Test results indicate that the Eagle 3000 Patient Monitor provides an equivalent level or better in performance, when compared to the legally marketed predicate devices when tested to the accuracy requirements as specified in the contents of the premarket notification submission."

Therefore, I cannot populate the table or answer questions related to specific performance metrics, sample sizes, ground truth establishment, or expert involvement as this information is not present in the provided text.

Based on the available information:

• 1. Table of acceptance criteria and the reported device performance:
  Cannot be provided as this information is not present in the 510(k) summary. The summary only states that the device provides "an equivalent level or better in performance" compared to predicate devices.

• 2. Sample sized used for the test set and the data provenance:
  Not specified in the provided text.

• 3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
  Not specified in the provided text.

• 4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:
  Not specified in the provided text.

• 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 monitoring system, not an AI-assisted diagnostic tool. No MRMC study or AI assistance is mentioned.

• 6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:
  Not applicable. This is a patient monitoring system, not an algorithm, and it explicitly states its use "by physicians, physician assistants, registered nurses, certified registered nurse anesthetists, or other hospital personnel trained in the use of the equipment," implying human interaction.

• 7. The type of ground truth used (expert consensus, pathology, outcomes data, etc):
  Not specified in the provided text. The summary only mentions "Precision and accuracy testing" against "accuracy requirements as specified in the contents of the premarket notification submission."

• 8. The sample size for the training set:
  Not applicable. This is a traditional patient monitoring system, and the concept of a "training set" for an algorithm is not mentioned.

• 9. How the ground truth for the training set was established:
  Not applicable, as there is no mention of a training set.

Ask a specific question about this device