Search Results

The IVD-GE02 system is a blood gases analyzer intended as an in vitro diagnostic device for the quantitative measurement of whole blood samples in a clinical laboratory. The IVD-GE02 system includes sensors for the measurement of pH, pCO2, pO2, potassium and glucose.

pH, pCO2, pO2: Measurement of blood gases (pCO2, pO2) and blood pH are used in the diagnosis and treatment of life-threatening acid-base disturbances.

Potassium: Measurement of potassium are used to monitor electrolyte balance in the diagnosis and treatment of disease conditions characterized by high or low potassium levels.

Glucose: Measurement of glucose is used in the diagnosis and treatment of carbohydrate metabolism disorders including diabetes mellitus, neonatal hypoglycemia, and idiopathetic hypoglycaemia and of pancreatic islet cell carcinoma.

IVD-GE02 calibration solutions are in vitro diagnostic products for the calibration of the IVD-GE02 test system for the measurements of pH, pCO2, pO2, potassium and glucose.

The IVD-GE02 system is an in-vitro diagnostic device for the determination of specified analytes in blood using electrochemical principles. The IVD-GE02 is essentially a modular system consists of the following components .-

• A dedicated combined PC and monitor with a touch screen .
• · A disposable cartridge containing
  • o Sensor
  • o Electronic circuitry, including an ASIC (Application Specific Integrated Circuit), to process sensor signals and drive the fluid management system.
• ◆ An instrument containing
  • o Fluid management components (tubing, pumps and valves)
  • o The cartridge (see above)
  • o An injection port for samples, calibration solutions and QC solutions
  • o Wiring to connect the cartridge to the fluid management components
• Syringes with calibration solutions packed in sealed foil bags .
• Containers with flush solutions ◆ ◆
• A waste container
• A dedicated printer and associated cables
• An ambient temperature and pressure sensor
• Cables to connect the monitor to the instrument and a power supply unit
• Packaging
• Label Copy

Here's a breakdown of the acceptance criteria and the study information for the IVD-GE02 device, as extracted from the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The document doesn't explicitly state quantitative acceptance criteria in terms of specific thresholds for R-squared values, standard deviations, or coefficients of variation that the device must meet. Instead, the acceptance is based on demonstrating "comparable performance" or "substantially equivalent" results to a predicate device (Rapidlab 865).

However, the provided performance data implies the target performance levels. The key performance metrics are linearity (R-squared), aqueous precision, and precision in blood (mean, standard deviation, and coefficient of variation), and method comparison (slope, intercept, R-squared).

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

• Linearity Study:
  • Sample Size: 7 levels of each analyte, read in duplicate, on each of three IVD-GE02 systems. The predicate device (Rapidlab 865) was also used.
  • Data Provenance: Human whole blood. The country of origin is not explicitly stated, but the submitter is based in the UK. The study was conducted following CLSI document EP6-A.
• Aqueous Precision Study:
  • Sample Size: Not explicitly stated for specific analytes, but the study design followed CLSI document EP5-A2, which typically involves repeat measurements over multiple days.
  • Data Provenance: Not specified, but "aqueous solutions" are used.
• Precision in Blood Study:
  • Sample Size: For each analyte, human blood samples were prepared at two different levels. Ten readings were taken at each analyte level.
  • Data Provenance: Human whole blood. Blood was drawn into heparinised vacutainers by trained phlebotomists. All blood was used within 24 hours of donation. Country of origin not explicitly stated.
• Method Comparison Study:
  • Sample Size:
    • pH: 104 samples
    • pCO2: 102 samples
    • pO2: 105 samples
    • K+: 111 samples
    • Glucose: 98 samples
  • Data Provenance: Human blood samples. The study was conducted in a "clinical setting." Data from the linearity study was also included to extend the analyte range. Country of origin not explicitly stated. The study followed CLSI guideline EP9A-2.
• Interferent Study:
  • Sample Size: Not explicitly stated, but various interferents at specific concentrations were tested for each analyte.
  • Data Provenance: Not specified for the test matrix (e.g., blood matrix or aqueous solution). The study design followed CLSI document EP7-A2.

All studies appear to be prospective, designed specifically for this validation.

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

This type of device (blood gas analyzer) relies on established analytical chemistry principles and reference methods, not expert human interpretation of images or complex clinical scenarios. Therefore, the concept of "experts establishing ground truth" in the way a radiologist does for an AI image analysis system is not directly applicable.

The "ground truth" for the test set is established by:

• The predicate device (Rapidlab 865), which is itself a legally marketed and established device for these measurements. The method comparison study directly compares the new device to the predicate.
• Reference materials/methods implied by the CLSI guidelines for linearity and precision.
• Known concentrations in calibration solutions and prepared blood samples.

No specific number or qualification of "experts" in the sense of human adjudicators for ground truth is mentioned because the ground truth is analytical measurement, not subjective interpretation.

4. Adjudication Method for the Test Set

Not applicable for a device that performs quantitative analytical measurements against a reference method or predicate device. There is no human "adjudication" in the traditional sense. The comparisons are statistical and direct measurements against the predicate.

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, an MRMC comparative effectiveness study was not done. This device is an in-vitro diagnostic instrument for direct measurement of blood analytes. It does not involve human "readers" or "AI assistance" in the interpretation of complex cases. Therefore, the concept of improvement in human readers with or without AI assistance is not relevant to this submission.

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

Yes, the studies described (Linearity, Aqueous Precision, Precision in Blood, Method Comparison, Interferent Study) represent the standalone performance of the IVD-GE02 device. While operators interact with the device to load samples and review results, the core measurement and data generation functionality is autonomous. There isn't an "algorithm only" performance that would be separated from the instrument's operation in this context, as the device itself is the measurement system. The performance metrics reported are for the device operating independently.

7. The Type of Ground Truth Used

The ground truth used for these studies is primarily:

• Reference measurements from the predicate device (Rapidlab 865): This is the direct comparison point for the method comparison study.
• Known concentrations of analytes: For linearity and precision studies, samples with established or prepared concentrations are used.
• Clinical laboratory standards and CLSI guidelines: These define the acceptable ranges and methodologies for establishing device performance.

It's based on objective analytical measurements, not expert consensus, pathology, or outcomes data in the usual sense.

8. The Sample Size for the Training Set

The document does not describe a traditional "training set" like one would find in a machine learning or AI development context. This device relies on electrochemical principles, not a machine learning model that requires a separate training phase. The development of the device's sensor characteristics and algorithms would have been an engineering process, not a data-driven training process in the AI sense.

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

As there is no explicit "training set" for an AI algorithm, this question is not directly applicable. The device's operational parameters and calibration are established through engineering design, internal testing, and the use of calibrated solutions as part of its normal operation (e.g., the IVD-GE02 calibration solutions mentioned).

Ask a specific question about this device