Search Results

The Maxtec MaxBlend 2+p is intended to provide a continuous air/oxygen gas mixture and to continuously monitor the concentration of oxygen and pressure being delivered to infant, pediatric, and adult patients. It is a restricted medical device intended for use by qualified, trained personnel, under the direction of a physician, in professional healthcare settings, i.e., hospital, subacute, and nursing-care facilities where the delivery and monitoring of air/oxygen mixtures is required. This is not intended as a life-supporting device or life sustaining device.

The Maxtec MaxBlend 2+p is an oxygen delivery device which incorporates an air/oxygen blender, battery powered oxygen and pressure monitor, and an adjustable flowmeter, all in a single assembly. The integral air/oxygen blender provides precise mixing of medical grade air and oxygen. The flowmeter provides control of the flow rate delivered. The oxygen monitor measures the oxygen concentration from the blender's gas flow, displays these measured concentrations, and provides user selectable high and low oxygen alarms. It also allows the user to monitor pressure simultaneously using adjustable high and low alarm limits.

The document describes the Maxtec MaxBlend 2+p, a medical device combining an air/oxygen blender, oxygen monitor, pressure monitor, and flowmeter. The 510(k) submission seeks substantial equivalence to existing predicate devices (Maxtec MaxBlend 2 - K161718 and MaxO2ME+p - K221734), particularly highlighting the addition of a pressure monitoring feature.

Here's an analysis of the acceptance criteria and the study information based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for several performance aspects are implicitly derived from the comparative tables (Table 1 and 2) where the subject device's specifications are listed and compared to the predicate devices. The "Reported Device Performance" for the subject device is simply its stated characteristics, which are claimed to meet or be identical to the predicate device specifications, thereby meeting the acceptance criteria based on substantial equivalence.

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

The document explicitly states:

• Non-clinical Testing: "We performed a number of tests to demonstrate that the proposed device performed as intended." It then lists several international standards that the device met. It does not specify a sample size for the test set or the data provenance (e.g., country of origin, retrospective/prospective). The testing appears to be primarily related to engineering and performance validation against standards rather than clinical data.
• Clinical Testing: "No human clinical testing was performed."

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

Since "No human clinical testing was performed," and the primary testing involved adherence to international standards and non-clinical performance, there is no mention of experts establishing ground truth for a clinical test set. The "ground truth" for the non-clinical testing would be the defined requirements and specifications within the cited standards.

4. Adjudication Method for the Test Set

Not applicable, as no human clinical test set is mentioned, and the testing focuses on compliance with technical standards.

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 medical gas mixer and monitor, not an AI-assisted diagnostic or interpretation tool that would involve human readers or image analysis.

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

The device is a hardware product with integrated monitoring capabilities. Its "standalone" performance is measured through its adherence to the technical specifications and international standards listed under "Non-clinical Testing." The document states: "In all cases the subject device met the standard requirements." This implicitly confirms standalone performance against established benchmarks.

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

For the performance specifications (e.g., oxygen accuracy, pressure accuracy, flow rates), the ground truth is based on metrological standards and calibrated equipment used in accordance with the listed international standards (e.g., ISO 80601-2-55, ISO 11195). For biocompatibility, the ground truth is established by adherence to ISO 10993 and ISO 18562 standards.

8. The Sample Size for the Training Set

Not applicable. This is a hardware medical device with integrated sensors and controls, not a machine learning or AI algorithm that requires a "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.

Ask a specific question about this device

The MaxO2 ME+p is an oxygen monitor with integrated pressure monitoring intended for continuous monitoring of the concentration of oxygen and pressure being delivered to patients ranging from newborns to adults. It can be used in the hospital and sub-acute settings. The MaxO2 ME+p is not intended as a life-supporting device or life sustaining device.

The MaxO2 ME +p is a battery powered oxygen and pressure monitor in a single assembly. The oxygen monitor measures the oxygen concentration from a gas source, displays these measured concentrations, and provides user selectable high and low oxygen alarms. It also the user to monitor pressure simultaneously and provides user selectable high and low pressure alarms.

The MaxO2 ME+p device is an oxygen monitor with integrated pressure monitoring. The acceptance criteria and supporting study details are as follows:

1. Table of Acceptance Criteria and Reported Device Performance:

The provided document compares the MaxO2 ME+p (proposed device) to two predicate devices: Maxtec - MaxO2 ME (K153659) for oxygen monitoring and Caradyne - Criterion 40 (K992101) for pressure monitoring.

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

• The document does not specify a sample size for a test set in terms of clinical data or patient samples.
• The study primarily relies on non-clinical testing (bench testing and adherence to standards) to demonstrate performance.
• No human clinical testing or animal testing was performed.
• The data provenance is not applicable as it's not a study involving patient data.

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

• This information is not provided in the document. Given that the testing was non-clinical (adherence to standards and bench testing), the "ground truth" would be established by the requirements of those standards and the accuracy of reference measurement equipment. Experts involved would be in engineering, quality assurance, and regulatory affairs, but specific numbers and qualifications are not detailed.

4. Adjudication method for the test set:

• An adjudication method is not applicable as there was no study involving human readers or interpretation of results that would require consensus among experts. The testing involved verifying the device's performance against defined technical specifications and industry standards.

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 comparative effectiveness study was done. This device is an oxygen and pressure monitor, not an AI-powered diagnostic imaging device or a system designed to assist human readers in interpretation. Therefore, this type of study is not relevant to the MaxO2 ME+p.

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

• This question is not directly applicable in the context of this device. The MaxO2 ME+p is a standalone physical device (monitor) that measures oxygen concentration and pressure. Its performance is evaluated through non-clinical bench testing against established standards and specifications, not through an "algorithm only" performance study in the way it might be for a diagnostic AI. The device's functioning is its standalone performance without human input beyond its operation.

7. The type of ground truth used:

• The "ground truth" for the performance evaluation was established by technical specifications, measurements from calibrated reference equipment, and compliance with recognized industry standards. These include:
  • AAMI ANSI ES 60601-1: Medical electrical equipment safety and essential performance.
  • IEC 60601-1-2: Electromagnetic Disturbances.
  • AIM Standard 7351731: Electromagnetic Immunity Test for Exposure to Radio Frequency Identification Readers.
  • IEC 60601-1-8: Alarm systems.
  • ISO 80601-2-55: Respiratory gas monitors.
  • ISO 10993-5, -10, -11, -18: Biocompatibility.
  • ISO 18562-2, -3: Particulate material and VOCs.

8. The sample size for the training set:

• This information is not applicable. This device is a hardware monitor, not a machine learning or AI algorithm that requires a training set of data.

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

• This information is not applicable, as there is no training set for this type of device.

Ask a specific question about this device

The Maxtec MaxCap Neo and MaxCap Ped are indicated to provide a semi-quantitative visualization of the CO2 in the patient airvay. It is an adjunct in patient assessment to be used in conjunction with other methods to determine clinical signs and symptoms by or on the order of a physician.

Patient Population: MaxCap Neo is intended for use with neonates and infants 250g to 6kg. MaxCap Ped is intended for use with pediatric patients 1kg to 15kg

Environment of Use: Hospital, sub-acute, pre-hospital, transport.

The proposed MaxCap Neo and MaxCap Ped are comprised of several components:

• . Housing with standard 15 mm / 22 mm fittings to connect to the ventilator assist device and a face mask or endotracheal tube
• . Colorimetric litmus media which has been treated with chemical to detect the presence of CO2 by a change in pH

The Maxtec MaxCap Neo and MaxCap Ped devices are semi-quantitative CO2 detectors that use a colorimetric, pH-sensitive dye to visualize CO2 in a patient's airway. The devices are intended for use as an adjunct in patient assessment.

Here's an analysis of the acceptance criteria and study information provided:

1. Table of Acceptance Criteria and Reported Device Performance:

The document doesn't explicitly state "acceptance criteria" for each performance metric, but rather lists specifications and performance data for the proposed devices and compares them to predicate and reference devices. We can infer the acceptance criteria from the context of these comparisons and discussions of "similar" or "lower" than predicates.

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

The document describes non-clinical performance testing that was conducted.

• Sample size: Not explicitly stated for each individual test. The statement "We have performed the following performance data and testing for the new models" implies that a sufficient number of devices were tested to evaluate each listed metric.
• Data provenance: Bench testing results performed by Maxtec, LLC. The document does not specify a country of origin for the data beyond that. It is retrospective in the sense that the testing has already been completed and is being reported as part of the 510(k) submission.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience):

This information is not provided as the device is a colorimetric CO2 detector. The "ground truth" for the performance tests (e.g., CO2 concentration, flow rate, internal volume) would be measured by laboratory equipment and calibrated standards, not by human experts interpreting clinical cases.

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

This information is not applicable as the described studies are non-clinical performance bench tests, not studies involving human interpretation or clinical 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:

This information is not applicable. The device is a physical, colorimetric CO2 indicator, not an AI-powered diagnostic tool. Therefore, no MRMC study or AI assistance evaluation was performed.

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

This information is not applicable. The device is a physical, colorimetric CO2 indicator, not an algorithm. The reported performance is based on the device's physical and chemical properties.

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

The ground truth for the non-clinical performance tests (e.g., color change accuracy, flow resistance, internal volume) was established using calibrated laboratory equipment and controlled gas mixtures to represent specific CO2 concentrations and flow rates. This is an objective measurement rather than expert consensus or pathology.

8. The sample size for the training set:

This information is not applicable. The device is a colorimetric indicator and does not involve machine learning or a training set.

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

This information is not applicable as there is no training set for this type of device.

Ask a specific question about this device

The MaxBlend 2 and MaxBlend Lite are designed to provide a continuous air/oxygen gas mixture and to continuously monitor the concentration of oxygen being delivered to infant, pediatric, and adult patients. It is a restricted medical device intended for use by qualified, trained personnel, under the direction of a physician, in professional healthcare settings, i.e.. hospital, sub-acute, and nursing-care facilities where the delivery and monitoring of air/oxygen mixtures is required. This is not intended as a life supporting device.

The MaxBlend family of blenders, MaxBlend 2 and MaxBlend Lite, are oxygen delivery devices which incorporate an air/oxygen blender, battery powered oxygen monitor, and an adjustable flowmeter, all in a single assembly. The air/oxygen blender provides precise mixing of medical grade air and oxygen. The flowmeter provides control of the flow rate delivered. The oxygen monitor measures the oxygen concentration from the blender's gas flow, displays these measured concentrations, and provides user selectable high and low oxygen alarms.

The MaxBlend 2 incorporates the air/oxygen blender within its enclosure. The MaxBlend Lite may be provided with or without the blender component pre-assembled, allowing the user to install the oxygen monitor/flowmeter module, the MaxBlend Lite component, on an existing compatible blender. The addition of the MaxBlend Lite module is intended to improve the safety of an existing air/oxygen blender that is in the user's possession. Both devices use the exact same oxygen monitor sensor and electronics, MaxO2ME (K153659). Both devices use the exact same components/assemblies in the flowmeter and sensor bleed manifold which form the gas pathway to the patient. The only substantive difference is the form of the enclosure for the monitor electronics.

This document is a 510(k) Premarket Notification from Maxtec, LLC, regarding their MaxBlend 2 and MaxBlend Lite devices. It aims to demonstrate substantial equivalence to a predicate device, the Bird Sentry Blender (K973646), and references other cleared devices (K153659 – Maxtec MaxO2ME, K883038 - CareFusion / Bird - Blender, K925982 - BioMed Devices - Blender).

Here's an analysis based on your request, highlighting that this document outlines non-clinical testing for a medical device that mixes and monitors breathing gases, not a typically AI-powered diagnostic imaging device. Therefore, many of the requested categories (like MRMC studies, roles of experts for ground truth, sample sizes for training/test sets for AI models, etc.) are not applicable or are addressed differently in the context of hardware device performance testing.

Acceptance Criteria and Device Performance (Non-Clinical/Hardware)

The acceptance criteria are primarily based on meeting the performance specifications of the predicate device and established standards for medical devices of this type. The "reported device performance" is generally stated as "passed" or "meets the acceptance criteria" for various tests.

1. Table of Acceptance Criteria and Reported Device Performance

Overall Conclusion found in the document: "In all cases the proposed device passed or meets the acceptance criteria. One can find the proposed device substantially equivalent to the predicate Bird Sentry Blender (K973646)."

Regarding the AI/Diagnostic-Specific Questions (Not Directly Applicable to this Hardware Device but Addressed for Completeness):

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

• Not Applicable in the AI/ML sense. This submission relates to physical device performance and safety testing. The "test set" would refer to devices manufactured and subjected to various non-clinical engineering and performance tests (e.g., electrical safety, EMC, accuracy, durability, environmental conditions, etc.). The sample size for these tests is not explicitly stated in terms of a "data set" but rather implies a sufficient number of representative devices for engineering validation.
• Data Provenance: The testing was conducted by or on behalf of Maxtec, LLC, in the USA (Salt Lake City, Utah). It's "prospective" in the sense that the new devices were built and tested to meet pre-defined criteria, not a retrospective analysis of patient data.

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

• Not Applicable in the AI/ML sense. Ground truth for device performance is established by engineering specifications, international and national standards (e.g., ISO, IEC, ANSI/AAMI), and comparison to a legally marketed predicate device.
• Qualifications of Experts: The experts are likely engineers, quality assurance personnel, and regulatory affairs professionals involved in device design, manufacturing, and testing, potentially with input from clinical users for usability studies. Their qualifications would be in engineering, quality management systems (QMS), and regulatory compliance for medical devices.

4. Adjudication Method for the Test Set:

• Not Applicable in the AI/ML sense. Device performance testing involves objective measurements against a standard or specification. There isn't an "adjudication" process like a clinical expert consensus on an image. Test results either pass or fail the predetermined criteria.

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

• No, not done. This is a hardware device focused on gas mixing and oxygen monitoring, not a diagnostic imaging AI algorithm that assists human readers. Therefore, an MRMC study is not relevant.

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

• Not applicable as an "algorithm only" study. The core components (oxygen monitor and blender) are "standalone" in function (they perform their tasks independently) and their performance was evaluated by integrating them into the MaxBlend family. The document states "The major components... have already been cleared for the identical indications for use, population, and environment of use. Integrating them into a single assembly for user convenience does not alter their performance."

7. The Type of Ground Truth Used:

• The "ground truth" for this device's performance is based on:
  • Engineering specifications and design requirements.
  • Performance characteristics of the legally marketed predicate device (Bird Sentry Blender - K973646).
  • Compliance with recognized industry standards (e.g., ISO, IEC, ANSI/AAMI).
  • Comparison to previously cleared reference devices (K153659 – Maxtec MaxO2ME, K883038 - CareFusion / Bird - Blender, K925982 - BioMed Devices - Blender) which contribute components to the MaxBlend family.
  • Objective physical measurements and test results (e.g., accuracy of oxygen control, response time, flow rates, pressure, etc.).

8. The Sample Size for the Training Set:

• Not Applicable in the AI/ML sense. There is no "training set" for an AI model. This is a hardware device.

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

• Not Applicable in the AI/ML sense. No training set exists.

Ask a specific question about this device

The MaxO2ME oxygen monitor is intended for continuous monitoring of the concentration of oxygen being delivered to patients ranging from newborns to adults. It can be used in the pre-hospital, hospital and sub-acute settings. The MaxO2ME is not intended as a life supporting device.

The MaxO-ME is a handheld oxygen analyzer/monitor capable of measuring the oxygen concentration from 0% to 100% in a sample gas. A MAX-550E oxygen sensor outputs a voltage which is used by the Max O2ME to determine the concentration of oxygen based on a calibration at room air or 100% oxygen. The MaxO2ME contains alarms that can be controlled by the user to set a maximum or minimum allowable oxygen concentration.

Below is an analysis of the provided text regarding the MaxO2ME oxygen monitor's acceptance criteria and studies.

1. Table of Acceptance Criteria and Reported Device Performance

The document outlines acceptance criteria implied by the "Performance Specifications" and the reported performance of the MaxO2ME device compared to a predicate device (Precision Medical PM5900 - K063096) and a reference device (MiniOX - K961644). The "Discussion" section for "Non-clinical Testing" states, "In all cases the proposed device passed or meets the acceptance criteria," but does not explicitly list the criteria themselves or the specific values for each test type in the Performance Testing section beyond the initial performance specifications table.

Based on the "Performance Specifications - The following is a list of the differences between the proposed device and the predicate" and the subsequent "Discussion" for performance, some criteria and performance points can be inferred:

Study Proving Acceptance Criteria:
The document states: "We performed a number of tests to demonstrate that the proposed device performed as intended."
The following non-clinical performance tests were conducted:

• ISO 80601-2-55 Performance of respiratory gas monitors
• ISTA2A Shipping Validation Test Report
• Sensor performance Test Report
• Gas leakage Test Report
• Interfering gas effects Test Report
• Temperature compensation Test Report
• Drift of measurement accuracy Test Report
• MaxO2ME Operating and Storage Environment Report
• Device Cleaning Report and Disinfection Test
• Measurement accuracy Test Report
• IEC 60601-1 Electrical safety
• IEC 60601-1-2 - EMC
• IEC 60601-1-8 - Alarms
• Shelf-life / Real-time

The document concludes: "In all cases the proposed device passed or meets the acceptance criteria."

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

The document does not specify the sample size for any test set or the data provenance. The tests listed are general performance and safety tests for a medical device, which typically involve device units rather than patient data.

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. The tests performed are objective, non-clinical engineering and performance tests on the device itself, not requiring expert ground truth for interpretation.

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

Not applicable. As noted above, the tests are objective, non-clinical performance and safety tests, not requiring adjudication of results from multiple reviewers.

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 is not an AI/imaging device. No MRMC study was conducted. The device is an oxygen monitor, and its performance tests are related to its physical and functional specifications.

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

Yes, the studies conducted were standalone tests of the device itself ("algorithm only" in the sense of the device's functionality) without human-in-the-loop performance being evaluated. The device is an oxygen monitor and its performance is measured against technical specifications.

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

The ground truth for the performance and safety tests would have been established by calibrated reference equipment and standardized testing protocols (e.g., ISO and IEC standards). For example, a gas analyzer would be calibrated against known gas concentrations to determine its accuracy.

8. The sample size for the training set

Not applicable. This is not a machine learning or AI-based device, so 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/machine learning model, this question is not relevant.

Ask a specific question about this device

The UltraMaxO2 Oxygen Analyzer is a tool used to measure oxygen purity, flow and pressure at the outlet of an oxygen concentrator. The UltraMaxO2 Oxygen Analyzer is intended to be used in an environment where oxygen concentrators are being serviced or repaired. This includes Hospitals, Nursing Homes, Extended Care Facilities, Patient Homes, and Respiratory Device Service and Repair Centers.

The UltraMaxO2 device is used for checking oxygen concentrator performance with the measure of oxygen purity, and flow at the outlet of an oxygen concentrator. The UltraMaxO2 Oxygen Analyzer functions by passing an ultrasonic pulse through the gas sample and measuring the amount of time required for the pulse to transit the sample chamber. The transit time is converted into a gas concentration via calibration data stored in the device. This can be done because the transit time varies according to the molecular mass of the gas in the chamber. Flow is determined using the difference between the ultrasonic pulse traveling against the flow and the ultrasonic pulse traveling with the flow. Pressure is measured using a separate pressure sensor that measures the pressure build-up in the device when the output port is blocked. The pressure is displayed in either kPa or PSI determined by a user operable switch in the battery compartment. The materials of the UltraMaxO2 include ABS plastic for the enclosures and the ultrasonic oxygen sensor tube body, adhesive backed Polycarbonate labels, PVC tubing and nickel plated brass fittings for the flow path, electronic circuitry including transducers/receivers, pressure, temperature and humidity sensors and an LCD screen.

The UltraMaxO2 Oxygen Analyzer is a device designed to measure oxygen purity, flow, and pressure from oxygen concentrators.

Here's a breakdown of its acceptance criteria and the study information based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria for the UltraMaxO2 Oxygen Analyzer are implicitly derived from the performance claims and comparison to predicate devices. The study demonstrates that the UltraMaxO2 meets or exceeds the performance of its predicates.

Study Proving Device Meets Acceptance Criteria:

The provided document refers to non-clinical functional and performance tests that were conducted to establish substantial equivalence. These tests aimed to demonstrate that the UltraMaxO2 performs as well as, or better than, the legally marketed predicate devices.

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

• Sample Size for Test Set: Not explicitly stated. The document mentions "non-clinical functional and performance tests" but does not detail the number of units tested or the number of measurements taken.
• Data Provenance: The tests were non-clinical, likely conducted in a laboratory or manufacturing environment. The country of origin of the data is not specified, but the submitter (Maxtec, LLC) is based in Salt Lake City, Utah, USA, implying the testing likely occurred in the USA. The data is prospective as it was generated specifically for the submission to demonstrate performance.

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

• Number of Experts: Not applicable. This was a non-clinical device performance test, not an evaluation requiring expert interpretation of results for ground truth. The 'ground truth' would have been established by reference standards or calibrated measurement equipment.
• Qualifications of Experts: N/A

4. Adjudication Method for the Test Set

• Adjudication Method: Not applicable. As this was a non-clinical performance test comparing measurements against known standards or predicate device specifications, an adjudication method for interpretations is not relevant.

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

• MRMC Study: No, an MRMC comparative effectiveness study was not done. The document explicitly states: "No clinical studies were performed for the UltraMaxO2 as the device represents a well known technology for a recognized indication as evidenced...by comparison to the predicate devices currently cleared for sale in the US market."
• Effect Size of Human Readers Improve with AI vs. without AI Assistance: Not applicable, as no MRMC or clinical study involving human readers or AI assistance was conducted.

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

• Standalone Performance: Yes, the fundamental performance evaluation was a standalone assessment of the device's accuracy in measuring oxygen purity, flow, and pressure. While not an 'algorithm-only' test in the AI sense, it represents the device's inherent capability independent of human interpretation. The "non-clinical functional and performance tests" assess the device's outputs directly against established standards or predicate device performance.

7. The Type of Ground Truth Used

• Type of Ground Truth: For the non-clinical functional and performance tests, the ground truth would have been established using calibrated reference standards or known gas mixtures/flow rates/pressures. For example, a known concentration of oxygen gas would be fed into the device, and the device's reading would be compared to that known concentration. Similarly, known flow rates and pressures would be applied. When comparing to predicate devices, the predicate device's established performance served as a benchmark for substantial equivalence.

8. The Sample Size for the Training Set

• Sample Size for Training Set: Not applicable. The UltraMaxO2 Oxygen Analyzer is a measurement device, not an AI/ML algorithm that requires a "training set" in the conventional sense. The device uses an ultrasonic pulse measurement method with "calibration data stored in the device." This calibration data would be established during manufacturing and testing, and is not a "training set" like those used for machine learning models.

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

• Ground Truth for Training Set: Not applicable in the AI/ML context. The device's internal calibration data is established by feeding known, precise inputs (e.g., specific oxygen concentrations, flow rates, pressures) into the device during its manufacturing and calibration process. The device's internal algorithms are then set to accurately report these known inputs. This would involve highly accurate laboratory equipment as the ground truth.

Ask a specific question about this device