Search Results

The COSMED K5 Wearable Metabolic Technology is a device designed to measure cardio-respiratory and metabolic functions during stress testing, rehabilitation, sports medicine and other related activities, performed in professional healthcare facilities only. The system is not intended for Home Use.

The main measurements reported by the system are Oxygen Uptake production, Ventilation, Heart Rate and Energy Expenditure.

K5 is intended to use with adults and children over the age of 14 years.

It is to be used by physicians or by trained personnel on a physician responsibility.

The system and its accessories are indicated for the acquisition, analysis, formatting, display, printing and storage of certain physiologic signals. It must not be intended as a monitoring device, nor as a sole means for determining a patient's diagnosis but for the purpose of assisting the clinician in the diagnosis of cardio-pulmonary diseases

The COSMED K5 Wearable Metabolic Technology is a portable unit, designed for the measurement of resting and exercise metabolism in the laboratory, utilizing international measurement quidelines published by relevant scientific societies. The K5 consists of the following main parts:

• Main portable unit
• Optional Bluetooth receiver
• PC software
• Face mask with head cap
• Turbine flowmeter with optoelectronic reader and wind cover
• Harness
• Heart rate monitor belt
• 02 sensor
• Battery charger + batteries
• AC/DC adapter
• USB cable
• Carrying case
• Optional pressure regulator for calibration

The K5 system is intended to be used the laboratory, by either direct connection or by telemetry. The system offers both 'breath-by-breath' and 'mixing chamber' operational modes.

Here's an analysis of the acceptance criteria and study information for the COSMED K5 Wearable Metabolic Technology, based on the provided FDA 510(k) summary:

Note: The provided document is an FDA 510(k) summary, which focuses on demonstrating substantial equivalence to a predicate device rather than providing a detailed clinical study report. Therefore, some of the requested information, particularly regarding specific study details like training set provenance, expert qualifications, and detailed statistical methodologies for clinical trials, is not fully elaborated in this type of document.

1. Table of Acceptance Criteria and Reported Device Performance

The document mentions "Accuracy validation (Vt, RF, VO2, VCO2)" as a non-clinical test. The table on page 7 provides specific measurement accuracies for ventilation, VO2, and VCO2, which serve as the acceptance criteria and reported performance.

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

The document refers to "non-clinical testing" for "Accuracy validation (Vt, RF, VO2, VCO2)." However, it does not specify the sample size used for these tests, nor does it provide details on the data provenance (e.g., country of origin, retrospective or prospective nature of data). These non-clinical tests are typically performed in a controlled laboratory environment using simulated physiological conditions or human volunteers, but the specifics are not detailed here.

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

This information is not provided in the 510(k) summary. The accuracy validation tests for metabolic devices often rely on comparison to a gold standard measurement method or calibrated gas mixtures, rather than expert consensus on patient data. Therefore, the concept of "experts establishing ground truth for the test set" in the context of this device's performance validation as presented is not applicable in the way it might be for image-based diagnostic AI.

4. Adjudication Method for the Test Set

This information is not provided and is generally not applicable to the type of non-clinical accuracy validation tests described for this device. Adjudication methods are typically used in clinical studies involving interpretation by multiple observers where there might be disagreement in assessing an outcome, which is not the case for direct physiological measurements.

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

No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done or at least not reported in this 510(k) summary. This type of study is usually conducted for diagnostic imaging devices where human readers interpret images with and without AI assistance to determine the AI's impact on diagnostic accuracy. The COSMED K5 is a measurement device for physiological variables, not an interpretative diagnostic imaging tool.

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

The performance data presented in the table above (Ventilation, VO2, VCO2 accuracy) represents the standalone performance of the device's measurement algorithms. The device directly measures and calculates these physiological parameters. There isn't an "algorithm only" vs. "human-in-the-loop" distinction in the same way there would be for an AI-powered diagnostic tool. The device is designed to provide direct physiological measurements, and the accuracy validation confirms its ability to do so.

7. The Type of Ground Truth Used

For the accuracy validation tests (Vt, RF, VO2, VCO2), the ground truth would typically be established by:

• Highly calibrated reference systems or gas analyzers: For VO2 and VCO2 measurements, this would involve comparing the device's readings against known gas concentrations or a highly accurate reference metabolic cart.
• Volumetric calibration for ventilation (Vt, RF): Using precision syringes or flow calibrators to provide known volumes and rates of airflow.

The document does not explicitly state the specific "gold standard" methods used, but for this type of device, it would be based on physical standards and calibrated instruments.

8. The Sample Size for the Training Set

The 510(k) summary does not provide information on a "training set" in the context of machine learning. This device measures physiological variables directly and calculates parameters based on known physiological equations and sensor readings. It's not an AI/ML device that requires a large dataset for training a predictive model in the conventional sense. The "training" of such a device would involve calibration and optimization of its internal algorithms and sensor performance based on engineering principles and validation data, not a "training set" of patient data for an AI model.

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

As there is no "training set" for an AI/ML model described for this device, the question of how its ground truth was established is not applicable in the AI sense. The development and calibration of the K5's measurement capabilities would rely on rigorous engineering, physical principles, and laboratory testing using established standards and known physiological conditions.

Ask a specific question about this device