Search Results

The ABL90 FLEX PLUS System is an in vitro diagnostic, portable, automated analyzer that quantitatively measures pH, blood gas (p02), Oximetry (s02, ctHb, FCOHb, FCOHb, FMetHb, and FHHb), in heparinized arterial and venous whole blood.

The ABL90 FLEX PLUS System is intended for use by trained technologists, nurses, physicians and therapists. It is intended for use in a laboratory environment, near patient, or point-of-care setting. These tests are only performed under a physician's order.

pH and pO2: pH and pO2 measurements are used in the diagnosis and treatment of life-threatening acid-base disturbances.

sO2: Oxygen saturation, more specifically the ratio between the concentration of oxyhemoglobin plus reduced hemoglobin.

ctHb (Total Hemoglobin): Total hemoglobin measure the hemoglobin content of whole blood for the detection of anemia.

FO2Hb: Oxyhemoglobin as a fraction of total hemoglobin.

FCOHb: Carboxyhemoglobin measurements are used to determine the carboxyhemoglobin content of human blood as an aid in the diagnosis of carbon monoxide poisoning.

FMetHb: Methemoglobin as a fraction of total hemoglobin.

FHHb: Reduced hemoglobin as a fraction of total hemoglobin.

The ABL90 FLEX PLUS System consists of the ABL90 FLEX PLUS analyzer, sensor cassette and solution pack consumables, and related accessories for the analyzers. The sensor cassettes, solution packs and related accessories are compatible with both analyzers. Multiple versions of the sensor cassettes are available. The sensor cassette versions vary in the maximum number of tests and availability of sensors for use. The solution pack is available in two versions, differing in the number of activities available.

The FDA 510(k) summary for the Radiometer ABL90 FLEX PLUS System provides detailed information about the device's analytical performance testing to demonstrate its substantial equivalence to its predicate device.

Here's a breakdown of the acceptance criteria and the study that proves the device meets them:

1. Table of Acceptance Criteria and Reported Device Performance

The acceptance criteria are implicitly defined by the reported performance metrics (linearity, detection limits, precision, bias) and the comparison to the predicate device. The goal is to show that the ABL90 FLEX PLUS System performs comparably to the predicate (ABL90 FLEX PLUS, K160153) and adheres to recognized standards (CLSI guidelines).

The summary does not explicitly present a table of "acceptance criteria" and "reported performance" side-by-side in a single formatted table. However, the various tables throughout the "Analytical Performance Testing Summary" section serve this purpose by presenting the measured performance of the ABL90 FLEX PLUS System against unstated, but implied, acceptable ranges or a comparison to the predicate.

Here's a synthesized representation of the reported device performance for key analytical parameters, effectively serving as the "reported device performance":

Key takeaway for "Acceptance Criteria": The general acceptance criterion for this 510(k) submission is to demonstrate "substantial equivalence" to the predicate device (ABL90 FLEX PLUS, K160153). While explicit numerical acceptance criteria are not presented in this summary document, the testing aims to show that the new device's performance (linearity, detection, precision, bias, interference) is comparable to the predicate and/or meets recognized clinical and analytical standards as outlined in CLSI guidelines.

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

• Linearity Testing: Numbers of samples are not explicitly stated for linearity testing, but it was conducted "in general accordance with CLSI EP06... and EP39."
• Detection Capability (LoB, LoD, LoQ): Numbers of samples are not explicitly stated.
• Precision (using stable, aqueous ampoule-based QC material):
  • N = 243 or 244 for each QC ampoule level and parameter.
  • Data Provenance: Testing occurred at three external sites. The specific countries of origin are not mentioned, but "external sites" suggests a multi-site study. This appears to be a prospective study, as it's part of the premarket submission.
• Precision (using blood):
  • N varies by parameter and test interval, ranging from 4 to 188 samples.
  • Data Provenance: Not explicitly stated, but implies collected from blood samples (human derived). Likely prospective data collected for the study.
• Method Comparison (Bias):
  • N varies by parameter, blood type, and mode (S65/SP65), ranging from 26 to 235 samples (arterial/venous blood).
  • Data Provenance: Not explicitly stated, but implies collected from patient blood samples. This would be prospective data collected specifically for the method comparison study.

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

This device (ABL90 FLEX PLUS System) is an in vitro diagnostic (IVD) analytical instrument. The ground truth for its performance is established through reference methods, defined concentrations of analytes in quality control materials, and comparison to a legally marketed predicate device, not typically through human expert adjudication of images or clinical outcomes that require multiple medical professionals.

Therefore, the concept of "experts establishing ground truth" in the way it might apply to an AI imaging device (e.g., radiologists reviewing scans) is not directly applicable here. The "experts" would be the laboratory personnel and analytical chemists who perform the testing and ensure adherence to CLSI guidelines. Their qualifications are implicitly assumed to be appropriate for performing such technical laboratory studies.

4. Adjudication Method for the Test Set

Not applicable for an IVD analytical instrument. Ground truth is established by reference methods, certified materials, and comparison with a predicate device, not by expert adjudication.

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

Not applicable. MRMC studies are typically performed for imaging devices or diagnostic aids where human interpretation is a key component, often comparing AI-assisted vs. unassisted human performance. This device is an automated, quantitative analytical instrument.

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

Yes, the performance data presented (linearity, detection/quantitation, precision, bias, interference) are all measures of the standalone analytical performance of the ABL90 FLEX PLUS System. The device provides quantitative measurements autonomously without continuous human interpretation required for each result.

7. The Type of Ground Truth Used

The ground truth used for this device's analytical performance studies are:

• Reference materials/Certified Analytes: For linearity, detection, and precision testing. These are materials with known, precisely measured concentrations of the analytes (pH, pO2, ctHb, sO2, etc.).
• Predicate Device Measurements: For method comparison/bias studies, the measurements from the legally marketed ABL90 FLEX PLUS (K160153) served as the comparator (or "ground truth" to determine bias relative to the predicate).
• CLSI Guidelines: The studies adhere to relevant Clinical and Laboratory Standards Institute (CLSI) guidelines (e.g., EP06, EP39, EP17-A2, EP05-A3, EP09c, EP07, EP37), which define accepted methodologies and performance characteristics for IVD devices.

8. The Sample Size for the Training Set

Not applicable. This document describes the performance testing for a finished IVD product, not the development or training of a machine learning model. IVD devices like the ABL90 FLEX PLUS System are based on established analytical principles (potentiometry, optical, spectrophotometry) and calibrated using defined reference materials, not "trained" on a dataset in the AI sense.

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

Not applicable, as there is no "training set" in the context of this device's analytical principles. Ground truth for calibration and development of such instruments is established through rigorous analytical chemistry methods using highly purified and characterized reference standards.

Ask a specific question about this device

The GEM Premier 7000 with iQM3 is a portable critical care system for use by health care professionals to rapidly analyze lithium heparinized whole blood samples at the point of health care delivery in a clinical setting and in a central laboratory. The instrument provides quantitative measurements of pH, pCO2, sodium, potassium, chloride, ionized calcium, glucose, lactate, hematocrit, total bilirubin, and CO-Oximetry (tHb, O2Hb, MetHb, HHb, sO2*) parameters from arterial, venous, or capillary lithium heparinized whole blood. These parameters, along with derived parameters, aid in the diagnosis of a patient's acid/base status, electrolyte and metabolite balance and oxygen delivery capacity.

*s02 = ratio between the concentration of oxyhemoglobin and oxyhemoglobin plus deoxyhemoglobin.

• · pH, pCO2, and pO2 measurements in whole blood are used in the diagnosis and treatment of life-threatening acid- base disturbances.
• · Electrolytes in the human body have multiple roles. Nearly all metabolic processes depend on or vary with electrolytes:
• Sodium (Na+) measurements are used in the diagnosis and treatment of aldosteronism, diabetes insividus, adrenal hypertension, Addison's disease, dehydration, inappropriate antidiuretic secretion, or other diseases involving electrolyte imbalance.
• Potassium (K+) measurements are used to monitor electrolyte balance in the diagnosis and treatment
• of disease conditions characterized by low or high blood potassium levels.
• Ionized calcium (Ca++) measurements are used in the diagnosis and treatment of parathyroid disease, a variety of bone diseases, chronic renal disease, and tetany.
• Chloride (Cl-) measurements are used in the diagnosis and treatment of electrolyte and metabolic disorders, such as cystic fibrosis and diabetic acidosis.
• · Hematocrit (Hct) measurements in whole blood of the packed red cell volume of a blood sample are used to distinguish normal from abnormal states, such as anemia and erythrocytosis (an increase in the number of red cells).
• · Glucose (Glu) measurement is used in the diagnosis, monitoring and treatment of carbohydrate metabolism
• disturbances including diabetes mellitus, neonatal hypoglycemia, idiopathic hypoglycemia, and pancreatic islet cell carcinoma.
• · Lactate (Lac) measurement is used:
• to evaluate the acid-base status of patients suspected of having lactic acidosis;
• to monitor tissue hypoxia and strenuous physical exertion;
• in the diagnosis of hyperlactatemia.
• · Total Bilirubin (tBili) measurement is used to aid in assessing the risk of kernicterus and hyperbilirubinemia in neonates.

• CO-Oximetry (tHb, COHb, MetHb, O2Hb, HHb, and sO2) evaluates the ability of the blood to carry oxygen by measuring total hemoglobin and determining the percentage of functional and dysfunctional hemoglobin species.

– Total Hemoglobin (tHb): Total hemoglobin measurements are used to measure the hemoglobin content of whole blood for the detection of anemia.

• COHo: Carboxyhemoglobin measurements are used to determine the carboxyhemoglobin content of human blood as an aid in the diagnosis of carbon monoxide poisoning.

• MetHb: Methemoglobin measurements are used to determine different conditions of methemoglobinemia.

• HHb: Deoxyhemoglobin, as a fraction of total hemoglobin, is used in combination with oxyhemoglobin to measure oxygen status.

• O2Hb: Oxyhemoglobin, as a fraction of total hemoglobin, is used in combination with deoxyhemoglobin to measure oxygen status.

• sO2: Oxygen saturation, more specifically the ratio between the concentration of oxyhemoglobin and oxyhemoglobin plus deoxyhemoglobin, is used to measure oxygen status.

The GEM Premier 7000 with iQMs system provides health care professionals with quantitative measurements of lithium heparinized whole blood pH, pCO2, pO2, Na*, K*, Ch, Ca**, glucose, lactate, Hct, total bilirubin and CO-Oximetry (tHb, O2Hb, COHb, MetHb, HHb, sO₂*) from arterial, venous or capillary samples at the point of health care delivery in a clinical setting and in a central laboratory.

*sO₂ = Ratio between the concentration of oxyhemoglobin plus deoxyhemoglobin plus deoxyhemoglobin.

Key Components:
Instrument: It employs a unique touch-sensitive color screen and a simple set of menus and buttons for user interaction. The analyzer guides operators through the sampling process with simple, clear messages and prompts.
PAK (Cartridge): All required components for sample analysis are contained in the GEM PAK, including sensors, optical cell for CO-Oximetry and total bilirubin, sampler, pump tubing, distribution valve, waste container and Process Control Solutions. The GEM PAK is an entirely closed analytical system. The operator cannot introduce changes to the analytical process before or during the GEM PAK's use-life on board the instrument. The GEM PAK has flexible menus and test volume options to assist facilities in maximizing efficiency. The EEPROM on the GEM PAK includes all solution values and controls the analyte menu and number of tests. The setup of the instrument consists of inserting the GEM PAK into the instrument. The instrument will perform an automated GEM PAK start-up during which the following is performed: warm-up (15 minutes), sensor conditioning (10 minutes), Process Control Solution (PCS) performance (15 minutes), all of which take about 40 minutes. After GEM PAK start-up, Auto PAK Validation (APV) process is automatically completed: two completely independent solutions traceable to NIST standards, CLSI procedures or internal standards, containing two levels of concentration for each analyte (PC Solution D and E), are run by the analyzer to validate the integrity of the PC Solutions and the overall performance of the analytical system. Note: GEM PAKs that include tBili analyte will require the successful performance of CVP 5 tBili. Includes all necessary components for hemolysis detection, such as an acoustofluidic flow cell, an LED light source and an optical detector, for appropriate flagging of potassium measurements in whole blood samples without additional sample volume or sample processing steps.
Intelligent Quality Management (iQM3): iQM3 is used as the quality control and assessment system for the GEM Premier 7000 system. iQM3 is an active quality process control program designed to provide continuous monitoring of the analytical process before, during and after sample measurement with real-time, automatic error detection, automatic correction of the system and automatic documentation of all corrective actions, replacing the use of traditional external QC. iQM3 introduces hemolysis detection in whole blood samples, enhancing quality assessment in the pre-analytical phase of testing.

Based on the provided text, the device in question is the GEM Premier 7000 with iQM3, which is a portable critical care system for analyzing blood samples. The document describes its comparison to a predicate device, the GEM Premier 5000, and discusses its performance studies.

Here's an analysis of the acceptance criteria and the study proving the device meets them:

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

The document does not provide a direct table of specific numerical acceptance criteria for each analyte's performance (e.g., pH, pCO2, Na+, etc.) nor does it list the reported device performance in those exact terms. Instead, it states that "All verification activities were performed in accordance to established plans and protocols and design control procedures. Testing verified that all acceptance criteria were met."

The "Performance Summary" section lists the types of studies conducted to demonstrate that the modifications (specifically the new iQM quality check/Hemolysis detection module) do not impact the performance data represented in the Operators Manual, aligning with recognized guidelines. This implies the acceptance criteria are tied to maintaining performance comparable to the predicate device and being within acceptable ranges as defined by the mentioned CLSI guidelines.

Therefore, a table of explicit numerical acceptance criteria and reported performance values for each analyte is NOT AVAILABLE in the provided text. The document broadly states that the device met its 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 mentions several types of performance studies:

• Verification (Internal Method Comparison, Internal Whole Blood Precision, Hemolysis Interference on Potassium, Hemolysis Verification)
• Shelf-life and Use-life studies

However, the specific sample sizes used for these test sets are NOT provided in the text. There is also no information about the data provenance (e.g., country of origin of the data, retrospective or prospective).

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

This information is NOT available in the provided text. The device is an in-vitro diagnostic (IVD) instrument that provides quantitative measurements of various blood parameters. The "ground truth" for such devices typically comes from reference methods, calibrated standards, or comparative analyses with established, highly accurate laboratory instruments, rather than human expert consensus on interpretations like with imaging.

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

Given that this is an IVD device for quantitative measurements of blood parameters, the concept of "adjudication" by multiple human readers (like in imaging studies) does not directly apply. Performance is assessed through analytical accuracy, precision, and interference studies against known standards or reference methods. Therefore, no adjudication method in the sense of expert consensus on interpretations is described or implied.

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

There is no indication that a multi-reader multi-case (MRMC) comparative effectiveness study was performed. This type of study is relevant for AI-assisted diagnostic tools where human interpretation is part of the workflow. The GEM Premier 7000 with iQM3 is described as an analytical instrument providing direct quantitative measurements, not an AI system assisting human readers with interpretation. The "iQM3" refers to Intelligent Quality Management, which is an automated quality control system for the instrument itself, not an AI for human diagnostic assistance.

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

The device itself is a standalone analytical instrument. The performance studies described (Internal Method Comparison, Internal Whole Blood Precision, Hemolysis Verification, etc.) essentially represent "standalone" performance, as they evaluate the accuracy and precision of the instrument's measurements directly. The iQM3 system is an internal quality control mechanism for the device's measurements. Therefore, yes, a standalone performance evaluation of the device's analytical capabilities was implicitly done.

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

For a device that provides quantitative measurements of blood parameters, the "ground truth" for the test set would typically be established using:

• Reference methods: Highly accurate and precise laboratory methods for measuring each analyte.
• Calibrated standards: Solutions with precisely known concentrations of the target analytes.
• Comparison to predicate device: As this is a 510(k) submission, a primary method of establishing "ground truth" performance for the new device is by comparing its measurements against those of a legally marketed predicate device (GEM Premier 5000), which itself would have been validated against reference methods and standards.

The text mentions "two completely independent solutions traceable to NIST standards, CLSI procedures or internal standards" for "Auto PAK Validation (APV)". This strongly suggests that traceable standards and potentially CLSI-defined reference methods were used to establish the ground truth for performance evaluation.

8. The sample size for the training set

The document describes the GEM Premier 7000 with iQM3 as a medical device for quantitative measurements, not explicitly as a machine learning/AI model that requires a "training set" in the conventional sense (i.e., for supervised learning). The iQM3 is an "active quality process control program" with "Pattern Recognition (PR) software." While pattern recognition might involve some form of "training" or calibration, the document does not specify a separate "training set" in terms of data volume for such a process. It focuses on the validation of the device's analytical performance. Therefore, the concept of a "training set" sample size as applicable to AI/ML devices is not explicitly discussed or provided.

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

As noted above, the primary function of GEM Premier 7000 with iQM3 is quantitative measurement. If the "iQM3" component involved training for its "Pattern Recognition (PR) software," the document does not detail how a specific ground truth for such training was established. It primarily discusses the use of "Process Control Solutions (PCS)" and "Calibration Valuation Product (CVP 5)" for system checks and validation ("Auto PAK Validation (APV) process"). These solutions, traceable to NIST or CLSI standards, function as internal reference points for the device's operational checks and quality control, which could be considered an ongoing form of "ground truth" to maintain analytical performance, rather than a one-time "training set" for model development.

Ask a specific question about this device

The GEM Premier 5000 is a portable critical care system for use by health care professionals to rapidly analyze heparinized whole blood samples at the point of health care delivery in a clinical setting and in a central laboratory. The instrument provides quantitative measurements of pH, pCO2, pO2, sodium, chloride, ionized calcium, glucose, lactate, hematocrit, total bilirubin and CO-Oximetry (tHb, O2Hb, COHb, MHb, sO2*) parameters from arterial, venous or capillary heparinized whole blood. These parameters, along with derived parameters, aid in the diagnosis of a patient's acid/base status, electrolyte and metabolite balance and oxygen delivery capacity.

*sO2 = ratio between the concentration of oxyhemoglobin plus deoxyhemoglobin plus deoxyhemoglobin.

· pH, pCO2, and pO2 measurements in whole blood are used in the diagnosis and treatment of life-threatening acid-base disturbances.

· Electrolytes in the human body have multiple roles. Nearly all metabolic processes depend on or vary with electrolytes:

· Sodium (Na+) measurements are used in the diagnosis and treatment of aldosteronism, diabetes insipidus, adrenal hypertension, Addison's disease, dehydration, inappropriate antidiuretic secretion, or other diseases involving electrolyte imbalance.

· Potassium (K+) measurements are used to monitor electrolyte balance in the diagnosis and treatment of disease conditions characterized by low or high blood potassium levels.

· Ionized calcium (Ca++) measurements are used in the diagnosis and treatment of parathyroid disease, a variety of bone diseases, chronic renal disease and tetany.

· Chloride (Cl-) measurements are used in the diagnosis and treatment of electrolyte and metabolic disorders, such as cystic fibrosis and diabetic acidosis.

· Hematocrit (Hct) measurements in whole blood of the packed red cell volume of a blood sample are used to distinguish normal from abnormal states, such as anemia and erythrocytosis (an increase in the number of red cells).

· Glucose (Glu) measurement is used in the diagnosis, monitoring and treatment of carbohydrate metabolism disturbances including diabetes mellitus, neonatal hypoglycemia, idiopathic hypoglycemia, and pancreatic islet cell carcinoma.

• · Lactate (Lac) measurement is used:
• · to evaluate the acid-base status of patients suspected of having lactic acidosis;
• · to monitor tissue hypoxia and strenuous physical exertion;
• in the diagnosis of hyperlactatemia.

· Total Bilirubin (tBili) measurement is used to aid in assessing the risk of kernicterus and hyperbilirubinemia in neonates.

· CO-Oximetry (tHb, COHb, MetHb, O2Hb. HHb, and sO2) evaluates the ability of the blood to carry oxygen by measuring total hemoglobin and determining the percentage of functional hemoglobin species.

• Total Hemoglobin (tHb): Total hemoglobin measurements are used to measure the hemoglobin content of whole blood for the detection of anemia.

· COHb: Carboxyhemoglobin measurements are used to determine the carboxyhemoglobin content of human blood as an aid in the diagnosis of carbon monoxide poisoning.

3

· MetHb: Methemoglobin measurements are used to determine different conditions of methemoglobinemia.

· HHb: Deoxyhemoglobin, as a fraction of total hemoglobin, is used in combination with oxyhemoglobin to measure oxygen status.

· O2Hb: Oxyhemoglobin, as a fraction of total hemoglobin, is used in combination with deoxyhemoglobin to measure oxygen status.

• sO2: Oxygen saturation, more specifically the ratio between the concentration of oxyhemoglobin and oxyhemoglobin plus deoxyhemoglobin, is used to measure oxygen status.

The GEM Premier 5000 system provides fast, accurate, quantitative measurements of heparinized whole blood pH, pCO2, pO2, Na+, K+, Cl-, Ca++, glucose, lactate, Hct, total bilirubin and CO-Oximetry (tHb, O2Hb, COHb, MetHb, HHb, sO2) from arterial, venous or capillary samples.

The provided text is a 510(k) summary for the GEM Premier 5000 device, detailing an operating system upgrade. This document is a regulatory submission for a device change and does not contain the information requested regarding acceptance criteria, device performance tables, study specifics (sample size, data provenance, expert qualifications, adjudication methods, MRMC studies, standalone performance), or ground truth establishment.

The submission is a Special 510(k), which indicates a modification to an already cleared device, not a de novo clearance requiring extensive clinical performance studies. The core of this submission is a software update (operating system change from Fedora 17 Linux to WindRiver LTS 18 Linux) with the stated reason to "accommodate long-term support of resolutions for common vulnerability exposures."

The document explicitly states:

• "Performance data is limited to Software Verification as the scope of this Special 510(k) is specific to an operating system upgrade..."
• "The changes in this submission do not introduce: Changes to indications for use or intended use, Changes to the fundamental scientific technology, Changes to operating principle, Changes to labeled performance claims."

Therefore, the requested information, which typically pertains to the establishment of initial clinical performance and effectiveness, is not present in this regulatory document for this specific submission. The focus here is on ensuring the device continues to meet its previously established performance claims after a technical software upgrade, rather than demonstrating new performance capabilities.

Ask a specific question about this device

The OPTI® B-Lac cassette is intended to be used for the in vitro measurement of pH, PO2, total hemoglobin (tHb), and % Saturated O2 in sodium heparinized venous blood samples on the OPTI CCA-TS and OPTI CCA-TS2 platform in a clinical laboratory location.
· Measurements of blood gases (pCO2, pO2) and blood pH are used in the diagnosis and treatment of life-threatening acid-base disturbances.

• · Total hemoglobin (tHb) measurement is used to determine the hemoglobin content of human blood.
  · Oxygen saturation (SO2) measurement is used to determine the oxygen capacity of the hemoglobin.

The OPTI CCA-TS/TS2 are portable devices, microprocessor-based instrument using optical fluorescence for the measurement blood gases, electrolytes and enzymes. The OPTI CCA-TS/TS2 utilize a color, graphical touch screen user interface. A disposable, single-use cassette contains all of the elements needed for calibration, sample measurement, and waste containment. Specific calibration from the cassette is scanned into the analyzer by holding the cassette package in front of the bar code scanner. The cassette is then placed into the measurement chamber. The analyzer warms the cassette to 37.0±0.1°C and performs a calibration verification. When calibration is verified, the analyzer aspirates the blood sample into the cassette and across the optode sensors. Fluorescence emission is then measured after equilibrating with the blood sample. After a single measurement, the cassette containing the blood sample is removed from the analyzer and discarded. The analyzer contains no reagents, blood, or waste. The B-Lac cassette is a disposable, single use cassette that contains four (4) sensors for in vitro quantitative measurements of PO2, PCO2, pH. There is an additional laser based measurement of total hemoglobin (tHb) and SO2. The B-Lac cassette is sealed in a foil pouch along with a desiccant and is marked with a barcode label that includes a lot identification number, calibration information, and expiration date.

The provided document describes the OPTI® B-Lac Cassette for in vitro measurement of blood gases and related parameters. Here's a breakdown of the requested information:

1. Table of Acceptance Criteria and Reported Device Performance

The document states that the performance of the redesigned B-Lac cassette was determined to meet the performance claims made in the original B-Lac cassette submission (K093280) for all analytes. However, the specific quantitative acceptance criteria from K093280 are not detailed in this document. The reported device performance is indicated by the statement that the device "meets the performance claims" or "was demonstrated to meet the performance claims."

Here's a table based on the information provided, noting where specific numerical criteria are not available in this document:

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

• 20-Day Precision:

  • Sample Size: Paired samples run twice daily over 20 days. Three lots of B-Lac cassettes were used, and three levels of aqueous quality control solution.
  • Data Provenance: In-house (presumably US-based, as the company is in Georgia, USA). Retrospective data analysis of prospective testing.

• Within-Run Precision:

  • Sample Size: Multiple repeats using three lots of B-Lac cassettes, three levels of aqueous quality controls, and whole blood manipulated to 3 different levels.
  • Data Provenance: In-house. Retrospective data analysis of prospective testing.

• Method Comparison (in-house):

  • Sample Size: Whole blood samples tonometered to different levels using different O2/CO2 gas mixtures to generate test levels for pH, PCO2, PO2, and SO2. Samples manipulated for tHb. The exact numerical count of samples or measurements is not specified.
  • Data Provenance: In-house. Retrospective data analysis of prospective testing.

• Method Comparison (Altitude):

  • Sample Size: Whole blood samples were tonometered to obtain samples that span the range for PCO2, PO2, and pH, and spiked or diluted for tHb. Aqueous solutions were measured. Number of samples/measurements not specified, but done at 4 distinct altitude sites (75 ft, 1080 ft, 5560 ft, 10151 ft).
  • Data Provenance: Conducted in the USA (Maine, Georgia, North Carolina, Colorado). Retrospective data analysis of prospective testing.

• Interference Testing:

  • Sample Size: 16 interferents tested for each analyte (PCO2, PO2, pH, tHb, SO2). The number of samples per interferent is not specified.
  • Data Provenance: In-house. Retrospective data analysis of prospective testing.

• Stability Testing:

  • Sample Size: Three lots of B-Lac cassettes were tested. One lot was subjected to two cycles of elevated and frozen temperatures.
  • Data Provenance: In-house. Retrospective data analysis of prospective testing for the initial 6 months, with real-time testing ongoing.

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

This document describes a medical device for in vitro diagnostic measurements. The "ground truth" for the test set is established by:

• Reference Methods/Predicate Devices:

  • Gravimetric target for PCO2 and PO2 (based on gas concentration).
  • Predicate device ABL90 Flex for pH, PCO2, PO2, and tHb.
  • E series cassettes on the OPTI CCA-TS/TS2 for SO2.

• No human "experts" (e.g., radiologists) were involved in establishing the ground truth in the way
  this question implies for imaging or subjective interpretation devices. The ground truth is
  based on established analytical methods and reference instruments.

4. Adjudication Method for the Test Set

Not applicable. This device provides quantitative measurements, and ground truth is established by reference methods/instruments, not through expert consensus requiring adjudication.

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

No. This document does not mention any MRMC comparative effectiveness study where human readers improve with or without AI assistance. This device is an in vitro diagnostic instrument, not an AI-assisted diagnostic tool for human interpretation.

6. Standalone Performance (Algorithm Only Without Human-in-the-Loop Performance)

Yes, the studies described are of standalone performance of the device (OPTI® B-Lac Cassette on OPTI CCA-TS/TS2 platforms). The measurements are performed automatically by the instrument and its embedded algorithms. There is no human-in-the-loop performance described or implied for the measurement process itself, although clinical interpretation of the results by healthcare professionals would follow.

7. Type of Ground Truth Used

The ground truth used for performance evaluation includes:

• Gravimetric targets: For PCO2 and PO2 (based on gas concentration for tonometered samples).
• Predicate device measurements: Measurements from the Radiometer ABL90 Flex for pH, PCO2, PO2, tHb, and from the OPTI CCA TS2 E-Series Cassettes for SO2.
• Aqueous quality control solutions and manipulated whole blood samples: Used for precision and linearity studies, where the expected values are known or derived from previous characterization.

8. Sample Size for the Training Set

The document does not explicitly mention a "training set" in the context of machine learning. The device utilizes "new algorithms utilized to calculate concentrations for these sensors" (specifically for PCO2) and has updated software. However, the data sets described are for performance verification and validation, not for training a new algorithm from scratch in the classical AI sense. If there was an internal dataset used for algorithm development or "training" (e.g., to derive calibration curves or correction factors), that information is not provided. The described studies are primarily for demonstrating post-development performance.

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

As no specific "training set" is described for algorithm development, the method of establishing ground truth for such a set is not provided. The document focuses on the verification and validation of the device's performance against established clinical and analytical standards.

Ask a specific question about this device

The HemoCue® Hb 301 System is intended for quantitative determination of hemoglobin in primary care or blood donation settings.

The HemoCue® Hb 301 System is intended to be used to determine the hemoglobin concentration for adults, adolescents, children, and infants above 1 month old in primary care setting.

The HemoCue® Hb 301 System is intended to be used to determine the hemoglobin concentration for adults in blood donation setting.

The HemoCue® Hb 301 System is for professional in vitro diagnostic use only.

The HemoCue® Hb 301 System provides a direct reading of the hemoglobin concentration in a sample using specially designed, single use microcuvette and an analyzer. The system can be used by non-laboratory personnel.

The HemoCue® Hb 301 System consists of the following parts:

• An analyzer supporting the following features:
  • Photometric determination of hemoglobin
  • Presentation of results on a display
• Power supply by power adapter or four AA batteries
• Single use microcuvettes (test consumable)
• Labeling:
  • Operating Manual
  • Package Insert
  • Quick reference Guide
  • Labels

The microcuvette serves both as a pipette and as a measuring cuvette. No dilution or other preparation of the blood sample is required before filling of the microcuvette. A whole blood sample of approximately 10 µL is drawn into the cavity in the microcuvette by capillary action.

The measurement takes place in the analyzer, which measures the absorbance of whole blood at an Hb/ HbO2 isosbestic point. The measurement is performed directly on the whole blood through measurement of the transmitted and scattered light and using an algorithm for translation into the hemoglobin concentration of the sample.

The HemoCue® Hb 301 System is traceable to the hemiglobincyanide (HiCN) method, the international reference method according to ICSH for the determination of the hemoglobin concentration in blood.

Here's the breakdown of the acceptance criteria and the study for the HemoCue® Hb 301 System, based on the provided document:

Acceptance Criteria and Device Performance

Study Details

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

   • Sample Size: 71 pediatric blood samples.
   • Data Provenance: Tested at one European clinical laboratory site. The data appears to be prospective as it describes a specific evaluation done to compare the device to the reference method and predicate device.

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

   • The document does not specify the number or qualifications of experts used to establish ground truth.

3. Adjudication method for the test set:

   • The document does not specify an adjudication method. The ground truth was established by a reference method (ICSH), which inherently has its own established protocol for measurement, rather than relying on expert consensus adjudication in this context.

4. 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 applicable. The HemoCue® Hb 301 System is an automated hemoglobin analysis device, not an AI-assisted diagnostic tool that human readers would interpret. Therefore, an MRMC study related to human reader improvement with AI is irrelevant to this device.

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

   • Yes, in essence. The study described is a direct comparison of the device's measurements (algorithmically determined hemoglobin concentration from photometric readings) against a reference method and a predicate device. The HemoCue® Hb 301 System itself performs the analysis without human interpretation of the final result. While "human-in-the-loop" isn't strictly defined for this type of device, the performance presented is of the automated system.

6. The type of ground truth used:

   • The primary ground truth used was the hemiglobincyanide (HiCN) method (ICSH reference method) for the determination of hemoglobin concentration.

7. The sample size for the training set:

   • The document does not explicitly state a sample size for a "training set" in the context of machine learning, as this device uses spectrophotometric measurements and an algorithm for translation into hemoglobin concentration, rather than a deep learning model that requires a distinct, large training set. The system is described as "factory calibrated."

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

   • The document states that "The HemoCue® Hb 301 System is traceable to the hemiglobincyanide (HiCN) method, the international reference method according to ICSH for the determination of the hemoglobin concentration in blood." This implies that the factory calibration (which is analogous to what might be considered a "training" or calibration phase for the device's internal algorithm) was established using the ICSH reference method as the gold standard. Specific details on the establishment of this ground truth for the factory calibration are not provided in this specific excerpt, beyond stating its traceability.

Ask a specific question about this device

Hemo Control is intended to be used for the quantitative determination of hemoglobin (Hb) concentrations in human blood.

The Hemo Control Hemoglobin Microcuvettes are intended to be used with the Hemo Control photometer for the quantitative determination of hemoglobin (Hb) concentrations in human blood.

For in-vitro diagnostic use only.

Hemo Control consists of the Hemo Control photometer / analyzer and the Hemo Control Hemoglobin Microcuvettes, its accessories and consumables (i.e. Control Solution Hb-con).
The Hemo Control photometer / analyzer is a semi-automated, spectrophotometric instrument, which provides instant quantitative total hemoglobin results.
Using the reagent filled microcuvette a small amount of arterial, venous or capillary blood is taken up by capillary action. The filled microcuvette is inserted into the Hemo Control photometer. The color produced by chemical reaction in the microcuvette is measured and the Hb value is displayed.
The measurement accuracy of the Hemo Control Hemoglobin Measurement System can be verified by use of Hb-con control solution, a quality control material with pre-determined hemoglobin concentration.
As a second quality control measurement, the control cuvette as a physical standard is used for a comfortable and cheap check of the device.

Here's a breakdown of the acceptance criteria and the study that proves the device meets them, based on the provided FDA 510(k) summary for the Hemo Control device:

Device: Hemo Control (automated hemoglobin system)

1. Table of Acceptance Criteria and Reported Device Performance

The document does not explicitly state "acceptance criteria" but rather presents performance characteristics of the device. The reported performance is the demonstration that the device meets the implicit acceptance criteria for substantial equivalence to its predicate device. For example, the low %CV values for precision demonstrate acceptable reproducibility. The regression equations with r-values close to 1 and slopes close to 1 for method comparison studies demonstrate agreement with reference methods and predicate devices.

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

• Precision: The document provides Coefficient of Variation (CV) values for "Within Run," "Total," and "Single Observation 20 days" at low (107 g/L), normal (129 g/L), and high (173 g/L) hemoglobin concentrations. The specific number of samples or measurements for each of these precision studies is not explicitly stated in this summary, but typically, precision studies involve repeated measurements of control materials or patient samples.
• Method Comparison (NCCLS Reference Method): n = 174 samples
• Method Comparison (HemoCue hemoglobin measurement system): n = 286 samples
• Method Comparison (Hemo Control Cuvettes in HemoCue): n = 286 samples
• Matrix Comparison (Capillary samples): n = 275 samples
• Matrix Comparison (Arterial samples): n = 10 samples

Data Provenance: The document does not explicitly state the country of origin or whether the data was retrospective or prospective. Given that this is a 510(k) submission from a German company (EKF-diagnostic GmbH) and involves comparison to established reference methods and other commercially available devices, it is highly likely these were prospective studies conducted in a clinical or laboratory setting.

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

This device is an automated hemoglobin system that measures a quantitative value (hemoglobin concentration). It does not rely on expert interpretation for its output. Therefore, the concept of "experts establishing ground truth" in the way it applies to image interpretation or diagnostic classification (e.g., radiologists) is not relevant here.

The "ground truth" for method comparison studies is established by:

• NCCLS Reference Method: This refers to a standardized laboratory method (likely cyanmethemoglobin method, which is a gold standard for hemoglobin measurement). The reference method itself is the "ground truth."
• Predicate Device (HemoCue hemoglobin measurement system): This is another commercially available, cleared device that serves as a comparison standard.

4. Adjudication Method for the Test Set

Not applicable. Since the device produces a quantitative numerical output for hemoglobin concentration, there is no need for an adjudication method among experts. The "ground truth" is determined by the reference method or comparison device.

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

No, an MRMC comparative effectiveness study was not done. This type of study is relevant for diagnostic devices where human readers interpret output (e.g., medical images) to assess the impact of AI assistance on their performance. The Hemo Control is an automated quantitative device.

6. If a Standalone (i.e., Algorithm Only Without Human-in-the-Loop Performance) Was Done

Yes, the analytical performance and method comparison studies described are inherently standalone performance studies. The Hemo Control device (photometer and microcuvettes) itself performs the measurement and provides the hemoglobin value. There is no "human-in-the-loop" interaction for interpreting the result, only for operating the device, taking the sample, and understanding the output.

7. The Type of Ground Truth Used

The ground truth used for performance evaluation included:

• NCCLS Reference Method: A highly standardized, accepted laboratory method for hemoglobin measurement.
• Predicate Device Performance: Comparison against the performance of another legally marketed device (HemoCue hemoglobin measurement system).

8. The Sample Size for the Training Set

The document does not explicitly mention a "training set" in the context of machine learning. This device operates based on spectrophotometric principles and chemical reactions, not on complex machine learning algorithms that require large training data sets in the typical sense. The fundamental physical and chemical principles are well-established.

The studies described (precision, linearity, method comparison, matrix comparison) are typically considered verification and validation studies performed after the device design is largely finalized, not for "training" an algorithm.

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

Not applicable, as a "training set" in the machine learning sense is not mentioned or implied for this device's operation. The device's calibration and performance are established against physical standards, reference methods, and quality control materials, which form its inherent "ground truth." Specifically, the device is calibrated against the NCCLS reference method, and its accuracy can be verified using Hb-con control solutions (quality control material with pre-determined hemoglobin concentration) and a control cuvette as a physical standard.

Ask a specific question about this device

The Stat Profile® Prime Plus Analyzer System is indicated for use by healthcare professionals in clinical laboratory settings and for point-of-care usage for quantitative determination of Hematocrit, Oxygen Saturation, Total Hemoglobin, Oxyhemoglobin, Carboxyhemoglobin, Methemoglobin, and Deoxyhemoglobin in heparinized arterial and venous whole blood.

The Stat Profile Prime Plus Analyzer System is a low cost, low maintenance analyzer for hospital laboratory and point-of-care settings. It consists of the analyzer, sensor cartridges, and thermal paper for an onboard printer. Optionally, it provides for reading of barcode labels (such as operator badges and data sheets).

The Stat Profile Prime Plus Analyzer has slots to accommodate two sensor cartridges (Primary and Auxiliary). The analyzer will determine the configuration of the system by detecting which sensor cards are installed.

As with the predicate, the Stat Profile Prime Plus Analyzer is a blood gas, co-oximetry, electrolyte, chemistry, and hematology analyzer with an enhanced test menu and multiple quality control options. Both traditional internal and external quality control is available, as well as an on-board Quality Management System (QMS), and an electronic monitoring approach that insures the analyzer is working properly at all times.

The Stat Profile Prime Plus Analyzer accepts samples from syringes and open tubes. The minimum sample size for analysis is 135 µL.

Sample collection, preparation and application to the analyzer are the same as for the previously cleared predicate. The end user can select which analytes are to be tested in the panel.

The provided text describes the performance validation of the Stat Profile® Prime Plus Analyzer System, specifically focusing on its point-of-care (POC) capabilities and comparability to a predicate device. This is a medical device, not an AI/ML software. Therefore, many of the requested categories related to AI/ML software development (e.g., number of experts for ground truth, adjudication method, MRMC studies, training set details) are not applicable to this document.

However, I can extract information related to the device's acceptance criteria and how its performance was proven.

Here's the breakdown of the available information:

Acceptance Criteria and Reported Device Performance

The acceptance criteria are implicitly defined by demonstrating "substantial equivalence" to the predicate device and by meeting established criteria for method comparison and imprecision, based on CLSI guidelines. The performance data is presented as method comparison statistics (slope, intercept, r-value) and total imprecision (SD, %CV).

Table of Acceptance Criteria and Reported Device Performance

The document doesn't explicitly state quantitative acceptance criteria thresholds for each parameter prior to the results. Instead, it presents the quantitative results from the study and concludes that these results "confirmed that the performance of the Stat Profile Prime Plus Analyzer System is substantially equivalent to that of the Nova Stat Profile pHOx Ultra Analyzer System (predicate device)."

Therefore, I will present the reported device performance, and the implicit acceptance can be inferred as meeting industry standards for substantial equivalence in medical devices of this type.

2. Sample Size and Data Provenance

• Test Set Sample Size:
  • Method Comparison (POC vs. Lab):
    • Hct: 417 samples
    • SO2: 398 samples
    • tHb: 416 samples
    • O2Hb: 422 samples
    • COHb: 425 samples
    • MetHb: 437 samples
    • HHb: 322 samples
  • Total Imprecision Performance: 20 runs performed on 3 analyzers, using 3 levels of quality control/linearity materials.
  • Within-Run Whole Blood Precision: Minimum of 2 operators per site across 3 POC sites (total 9 operators). Each precision run consisted of 10 replicate measurements using 7 different whole blood samples (5 native, 2 altered).
• Data Provenance: The study was conducted in the United States, across three Point-of-Care (POC) sites: a Cardiothoracic Intensive Care Unit (CTICU), an Emergency Department (ED), and a Respiratory Therapy Lab (RT). The data utilized a mix of quality control materials and discarded blood gas specimens. This suggests prospective collection of real-world samples within a clinical setting.

3. Number of Experts and Qualifications for Ground Truth

• This device is a medical diagnostic instrument, not an AI/ML algorithm. Ground truth for the method comparison study was established by laboratory measurements (presumably using the established predicate device or a gold standard lab method) which the new device's results were compared against.
• The study involved "trained Healthcare Professionals" and "POC personnel". These personnel performed the tests on the new device. Their qualifications are described as "trained, qualified staff found in typical POC sites where blood gas analyzers are utilized," including Respiratory Care, Nursing, and Exercise Physiology personnel. No "experts" in the sense of adjudicating image interpretations are mentioned, as this is not an imaging AI.

4. Adjudication Method for the Test Set

• Not applicable. This is a quantitative measurement device, not an interpretive one requiring expert adjudication. The comparison was quantitative against a reference method.

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

• Not applicable. This is a quantitative measurement device, not an AI-assisted diagnostic tool that would be evaluated for human reader improvement. The study compares the new device's performance to a predicate device and laboratory methods.

6. Standalone Performance

• Yes, effectively. The "Stat Profile® Prime Plus Analyzer System" provides quantitative measurements. The performance data presented (Method Comparison, Total Imprecision, Within-Run Whole Blood Precision) represents the standalone performance of the device in generating these measurements, compared against established laboratory methods or statistical precision targets. There is no "human-in-the-loop" performance as the human simply operates the device to obtain the measurement.

7. Type of Ground Truth Used

• Comparative ("Reference") Method: For the method comparison study, the ground truth was the results obtained from analyses performed by "Lab" (laboratory reference methods, presumably the predicate device or another validated laboratory analyzer). This is a comparative ground truth against an established standard.
• Statistical Targets: For precision studies, the ground truth is implicitly defined by the acceptable statistical variance and bias from the mean of repeated measurements, often against known control material values.

8. Sample Size for the Training Set

• Not applicable. This is a physical, chemical, and optical measurement device, not a machine learning model that requires a "training set" in the computational sense. The device's measurement algorithms are fixed based on spectrophotometric and impedance principles.

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

• Not applicable, as there is no "training set" for an AI/ML model for this device. The principles of measurement are based on established scientific methods (e.g., spectrophotometry and impedance) and not trained on data in the AI/ML sense.

Ask a specific question about this device

The RAPIDPoint® 500e Blood Gas System is in vitro diagnostic use and is designed to provide the determination in whole blood for the following parameters:

• Partial pressure of carbon dioxide ●
• Partial pressure of oxygen
• pH
• Sodium ●
• Potassium
• lonized Calcium ●
• Chloride .
• Glucose ●
• . Lactate
• . Total Hemoglobin and fractions: FO2Hb, FCOHb, FMetHb, FHHb
• . Neonatal Bilirubin

The RAPIDPoint 500e Blood Gas System is also intended for in vitro testing of pleural fluid samples for the pH measurement of pleural fluid can be a clinically useful tool in the management of patients with parapneumonic effusions.

The following critical value applies to pleural fluid pH > 7.3 is measured in uncomplicated parapneumonic effusions. All pleural fluids with a pH measurement

The RAPIDPoint 500e Blood Gas System is a compact, bench-top analyzer designed for in vitro diagnostic testing and is suitable for professional use in a point-of-care or central laboratory environment. This system measures the following: blood gases, electrolytes, total hemoglobin, and hemoglobin derivatives in arterial, venous, and capillary whole blood samples. Additionally, the RAPIDPoint 500e Blood Gas System measures pH in pleural fluid.

The RAPIDPoint 500e Blood Gas System incorporates a cartridge-based design with no external reagent bottles or gas tanks. The system uses self-contained measurement and wash/waste cartridges that are replaced when depleted. The system automatically calibrates the measurement sensors and reports results within 60 seconds for display on a color touch screen for easy viewing.

The provided text concerns the FDA 510(k) summary for the RAPIDPoint® 500e Blood Gas System. This document describes a modification to an existing device (RAPIDPoint 500 System) and asserts its substantial equivalence for FDA clearance. Therefore, a study to prove the device meets specific acceptance criteria for a new clinical claim (such as disease diagnosis accuracy based on images) is not applicable here.

The submission is a Special 510(k), meaning the changes are minor and do not alter the intended use, fundamental scientific technology, labeling, or principle of operation. The primary changes are an operating system update and minor hardware/software enhancements.

Key points from the document regarding "acceptance criteria" and "study:

• No new performance claims: The document explicitly states: "There is no change to labeled performance claims." This implies that the performance criteria previously established for the predicate device (RAPIDPoint 500 System) are still considered valid and met by the modified device.
• Verification and validation activities: The document states: "All verification and validation activities were performed in accordance to relevant standards, established plans and protocols and Siemens Design Control procedures. Testing verified all acceptance criteria were met." This refers to internal engineering and design control testing to ensure the modifications haven't negatively impacted the known performance characteristics of the device.

Given this context, I cannot generate a table of acceptance criteria and reported device performance in the typical sense of a clinical diagnostic study with new performance endpoints. The "acceptance criteria" here refers to demonstrating that the modified device performs comparably to the predicate for all existing measurements and that the new software/hardware features function as intended without compromising safety or effectiveness.

Therefore, the requested information elements cannot be fully addressed in the way they would for a device making new diagnostic claims based on a primary clinical study.

Here's an attempt to answer the questions based on the available information, noting where specific details are not provided:

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

   As this is a Special 510(k) for a modified device with "no change to labeled performance claims" and "no change to principle of operation," there isn't a new set of clinical performance acceptance criteria and reported performance figures presented in this summary document. The acceptance criteria implicitly refer to demonstrating that the modified device's performance is substantially equivalent to the predicate device across all analytes and clinical uses, meaning it continues to meet the predicate's established performance specifications. The document states: "Performance testing results were also comparable."

   Inferred Acceptance Criteria (based on substantial equivalence to predicate):
   The modified device (RAPIDPoint 500e Blood Gas System) must demonstrate performance comparable to the legally marketed predicate device (RAPIDPoint 500 System) for all measured analytes (pCO2, pO2, pH, Sodium, Potassium, Ionized Calcium, Chloride, Glucose, Lactate, Total Hemoglobin and fractions, Neonatal Bilirubin) in whole blood and pH in pleural fluid. This comparability would typically be assessed by demonstrating agreement (e.g., bias, precision, linearity) within acceptable limits as defined for the predicate device.

   Reported Device Performance:
   The document states: "Performance testing results were also comparable." Specific numerical performance data (e.g., accuracy, precision) for each analyte for the modified device are not provided in this 510(k) summary, as the submission focuses on substantial equivalence of the modified device to the predicate, rather than establishing new performance specifications.

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

   Not explicitly stated in the provided document. The document refers to "verification and validation activities" and "performance testing results," but does not detail the sample sizes or the provenance (e.g., country of origin, retrospective/prospective) of the data used for establishing "comparable" performance. Given it's a diagnostic device for blood analysis, samples would typically be human blood and pleural fluid.

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

   Not applicable or not specified. For laboratory diagnostic devices like blood gas analyzers, "ground truth" is typically established by reference methods or highly accurate laboratory instruments, not by a panel of human experts interpreting data.

4. Adjudication method for the test set

   Not applicable or not specified. Adjudication methods (e.g., 2+1, 3+1) are typically used in clinical studies for endpoint determination (e.g., presence of disease from images) based on multiple expert opinions. For a blood gas system, the performance is evaluated by comparing measurements against reference methods, not subjective 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

   Not applicable. This device is an in vitro diagnostic (IVD) blood gas system, not an AI-assisted diagnostic imaging or interpretation device that involves human "readers."

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

   This device is an automated in vitro diagnostic system. Its "standalone" performance means the accuracy and reliability of its measurements of various blood parameters. While the document mentions software changes and cybersecurity enhancements, it is not an algorithm that interprets human-generated data or makes a diagnosis. Its performance is inherent to its measurement capabilities. The 510(k) summary asserts that its performance is "substantially equivalent" to the predicate, implying successful internal testing to confirm this.

7. The type of ground truth used

   For IVD devices measuring physiological parameters, "ground truth" is typically established by:

   • Reference methods: Highly accurate and validated analytical methods (e.g., gas chromatography for blood gases, gravimetric methods for electrolytes) or
   • Comparison to predicate/established devices: Comparing measurements from the device under evaluation to a legally marketed and well-characterized predicate device or other established clinical laboratory instruments.

   This document explicitly states the modified device is "substantially equivalent to the comparative method" (referring to the predicate device).

8. The sample size for the training set

   Not applicable or not specified. This is a measurement device for chemical and physiological parameters, not a machine learning model that requires a distinct "training set" of clinical data in the typical sense. Any internal developmental data would be for engineering optimization rather than model training.

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

   Not applicable (as it's not an AI/ML device requiring a training set in that context).

Ask a specific question about this device

The HemoCue® Hb 801 System is intended for the quantitative determination of hemoglobin in capillary or venous whole blood (K2EDTA and Li-Heparin) in point-of-care settings. The HemoCue® Hb 801 System is intended to be used to determine the hemoglobin concentration for adults, adolescents, children, and infants above 1 month old. The HemoCue® Hb 801 System is for professional in vitro diagnostic use only.

The HemoCue® Hb 801 System provides a direct reading of the hemoglobin concentration in a sample using specially designed, single use microcuvette and an analyzer. The system can be used by non-laboratory personnel.

The HemoCue® Hb 801 System consists of the following parts:

• An analyzer supporting the following features: .
  • O Photometric determination of hemoglobin
  • Presentation of results on a display O
  • O Wired and wireless communication (USB and Bluetooth)
• Power supply by power adapter, chargeable or non- chargeable batteries ●
• Single use microcuvettes (test consumable)
• Labeling: ●
  • O Operating Manual
  • o Package Insert
  • Quick reference Guide o
  • o Labels

The microcuvette serves both as a pipette and as a measuring cuvette. No dilution or other preparation of the blood sample is required before filling of the microcuvette. A whole blood sample of approximately 10 uL is drawn into the cavity in the microcuvette by capillary action.

The measurement takes place in the analyzer, which measures the absorbance of whole blood at an Hb/ HbO2 isosbestic point. The measurement is performed directly on the whole blood through measurement of the transmitted and scattered light and using an algorithm for translation into the hemoglobin concentration of the sample.

The HemoCue® Hb 801 System is traceable to the hemiglobincyanide (HiCN) method, the international reference method according to ICSH for the determination of the hemoglobin concentration in blood.

Here's a summary of the acceptance criteria and study details for the HemoCue® Hb 801 System based on the provided text:

1. Table of Acceptance Criteria and Reported Device Performance

The document largely focuses on demonstrating equivalence to a predicate device and established reference methods, with direct explicit acceptance criteria only mentioned for linearity and anticoagulants. Performance is reported in terms of precision, linearity, and correlation with reference methods.

• Hb 2.0-3.0 g/dL: Mean 2.43 g/dL, SD 0.05
• Hb 6.0-7.0 g/dL: Mean 6.55 g/dL, SD 0.07
• Hb 9.5-10.5 g/dL: Mean 9.96 g/dL, CV 0.68% (Repeatability), 0.71% (Within Lab), 1.11% (Reproducibility)
• Hb 13.5-14.5 g/dL: Mean 14.07 g/dL, CV 0.71% (Repeatability), 0.82% (Within Lab), 1.16% (Reproducibility)
• Hb 16.5-17.0 g/dL: Mean 16.87 g/dL, CV 0.60% (Repeatability), 0.73% (Within Lab), 0.95% (Reproducibility)
• Hb 23.0-24.0 g/dL: Mean 23.39 g/dL, CV 0.67% (Repeatability), 0.77% (Within Lab), 0.97% (Reproducibility)

Overall precision (Multi-site):

• Hb 2.0-3.0 g/dL: Mean 2.30 g/dL, SD 0.03 (R), 0.04 (WL), 0.05 (R)
• Hb 6.0-7.0 g/dL: Mean 6.55 g/dL, SD 0.07 (R), 0.07 (WL), 0.08 (R)
• Hb 9.5-10.5 g/dL: Mean 10.24 g/dL, CV 1.04% (R), 1.17% (WL), 1.63% (R)
• Hb 13.5-14.5 g/dL: Mean 13.91 g/dL, CV 0.71% (R), 0.75% (WL), 1.00% (R)
• Hb 16.5-17.0 g/dL: Mean 16.75 g/dL, CV 0.52% (R), 0.63% (WL), 0.66% (R)
• Hb 23.0-24.0 g/dL: Mean 23.35 g/dL, CV 0.74% (R), 0.74% (WL), 0.89% (R) |
  | Precision (Quality Control) | SD and CV calculated for repeatability, between-run, between-day, and within laboratory precision for each level were within the defined acceptance criteria. | Overall precision (Multi-site, multi-lots, operators, days): All reported SD and CV values (e.g., Low control: Mean 6.34 g/dL, SD 0.05 (R), 0.04 (WL), 0.06 (R)) were within the defined acceptance criteria. |
  | Linearity | "fulfilled acceptance criteria for the non-linear error" (for range 1.0-25.6 g/dL). | System determined to be linear in the range 1.0-25.6 g/dL. |
  | Detection Limit (LoB) | Not explicitly stated (calculated to be 0.26 g/dL). | LoB for the HemoCue® Hb 801 System was determined to be 0.26 g/dL. |
  | Detection Limit (LoD) | Not explicitly stated (calculated to be 0.3 g/dL). | LoD for the HemoCue® Hb 801 System was determined to be 0.3 g/dL. |
  | Quantification Limit (LoQ) | Total Error (TE) for each sample ≤ 0.5 g/dL. | LoQ for the HemoCue® Hb 801 System was determined to be 0.5 g/dL, meeting the TE goal. |
  | Analytical Specificity | No significant interference at tested concentrations. | Most tested substances (Acetaminophen, Ascorbic acid, Creatinine, HbCO, HbO2, Hemolysis, Ibuprofen, MetHb, Platelets, Total protein, Salicylic acid, Simvastatin, Tetracycline, Triglycerides, Urea, Uric acid, Warfarin, Normal blood pH) showed no interference at respective concentrations.

Interference observed with:

• Conjugated bilirubin (>23 mg/dL at 10 g/dL Hb)
• Unconjugated bilirubin (>12 mg/dL at 10 g/dL Hb, >23 mg/dL at 20 g/dL Hb)
• Intralipid (>214 mg/dL at 10 g/dL Hb, >483 mg/dL at 20 g/dL Hb)
• Leucocytes (>260 x 10^9/L at 6.8 – 14.7 g/dL Hb)
  (Note: "May give elevated results in higher substance concentrations".) |
  | Anticoagulant Equivalence | Met the acceptance criteria regarding the correlation and bias between K2EDTA and Li-Heparin. | Both K2EDTA and Lithium Heparin samples fulfilled the acceptance criteria. |
  | Capillary vs. Venous Sample | Met the defined acceptance criteria and showed equivalency. | Both venous and capillary samples were within the defined acceptance criteria and showed equivalency. |
  | Method Comparison (Predicate)| Linear regression analysis demonstrated comparable performance. | Venous (N=264): Slope 1.00, Intercept -0.14, r 1.00 against predicate.
  Capillary (N=233): Slope 1.07, Intercept -0.91, r 0.96 against predicate.

Pediatric samples (N=71):

• HemoCue® Hb 801 vs ICSH: slope 0.95, r 0.99.
• HemoCue® Hb 801 vs HemoCue® Hb 301 System: slope 0.97, r 0.99.

Both samples types were within defined acceptance criteria. |
| Reference Range Verification| Values fall within the expected pediatric and adult reference intervals. | Study results verified that the reference ranges data on the HemoCue® Hb 801 System for the subgroups fall within the defined reference ranges (e.g., Infant 9.4-14.1 g/dL, Adult Male 13.0-17.0 g/dL). |

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

• Precision (Whole Blood):

  • Multi-microcuvette lots study: 6 hemoglobin levels, 5 operating days, 5 replicates per run, 3 microcuvette lots. Total 75 measurements per level. (Implied N = 6 levels * 75 measurements = 450).
  • Multi-site study: 3 sites, 5 operating days, 5 replicates per run, 6 hemoglobin levels. Total 75 measurements per level. (Implied N = 6 levels * 75 measurements = 450).
  • Provenance: Not explicitly stated, but clinical studies for method comparison mention primary care settings in the US and one European clinical laboratory. It is likely these precision studies were conducted in similar clinical laboratory environments. Retrospective/Prospective not specified, but likely prospective.

• Precision (Quality Control Material):

  • Sample size: Three sites, 20 operating days, 1 lot of control material (3 levels), duplicate runs twice daily. Total 240 measurements per level (80 per site). (Implied N = 3 levels * 240 measurements = 720).
  • Provenance: Not explicitly stated, but likely clinical laboratory settings, potentially those mentioned for method comparison (US/Europe). Likely prospective.

• Linearity:

  • Sample size: 9 hemoglobin concentrations, 15 replicates per concentration (5 replicates/analyzer). (Implied N = 9 concentrations * 15 replicates = 135).
  • Provenance: Clinical laboratory, likely prospective.

• Detection Limit (LoB, LoD, LoQ):

  • LoB: 4 individual plasma samples. 3 analyzers, 2 microcuvette lots, 4 operating days. 1 sample analyzed each day, 3 runs per day, 2 replicates/analyzer per run. Total 72 replicates/microcuvette lot (Implied N = 2 lots * 72 replicates = 144).
  • LoD: 4 K2EDTA whole blood samples from different subjects. 3 analyzers, 2 microcuvette lots, 4 operating days. 1 sample analyzed each day, 3 runs per day, 2 replicates/analyzer per run. Total 72 replicates/microcuvette lot (Implied N = 2 lots * 72 replicates = 144).
  • LoQ: 4 K2EDTA whole blood samples from different subjects. 3 analyzers, 2 microcuvette lots, 4 operating days. 1 sample analyzed each day, 3 runs per day, 3 replicates/analyzer per run. Total 108 replicates/microcuvette lot (Implied N = 2 lots * 108 replicates = 216).
  • Provenance: Clinical laboratory, likely prospective.

• Analytical Specificity:

  • Sample size: K2EDTA whole blood samples with adjusted Hb levels. Number of samples/subjects not explicitly stated for each interferent, but tested at two Hb concentrations (10±0.5 and 20±1.0 g/dL).
  • Provenance: Clinical laboratory, likely prospective.

• Anticoagulant Equivalence:

  • Sample size: 120 subjects provided both K2EDTA and Li-Heparin venous whole blood. Additional 11 subjects contributed samples to adjust Hb values.
  • Provenance: Two sites, likely clinical settings, likely prospective.

• Capillary vs. Venous Sample Equivalence:

  • Sample size: 40 subjects for direct comparison, plus 212 subjects from the method comparison study (total 252).
  • Provenance: Not explicitly stated, but likely related to the multi-site method comparison study in the US. Likely prospective.

• Method Comparison (Predicate):

  • Sample size:
    • US Study: 264 venous samples (28 contrived) and 233 capillary samples. Total 497.
    • European Clinical Lab: 71 pediatric samples.
  • Provenance:
    • US Study: Five primary care settings in the US.
    • European Clinical Lab Study: One European clinical laboratory site.
  • Retrospective/Prospective: Likely prospective.

• Reference Range Verification:

  • Sample size: Whole blood specimens collected. Number of individual samples not specified.
  • Provenance: 5 point-of-care locations in the US. Likely prospective.

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

• No information provided regarding experts establishing ground truth for the test set. Most studies compare the device against a predicate device or the International Council for Standardization in Haematology (ICSH) reference method, which represents a gold standard, not expert consensus.

4. Adjudication Method for the Test Set

• Not applicable/Not mentioned. The studies described are analytical performance studies comparing the device to reference methods or a predicate, not studies involving human interpretation or adjudication of results. For the method comparison studies, duplicates were used for the predicate device, and triplicates for the ICSH reference method, implying a direct comparison of numerical results rather than an adjudication process.

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

• No, an MRMC comparative effectiveness study was not done. The studies are focused on the analytical performance of the device in measuring hemoglobin concentrations, primarily comparing it to a predicate device and a reference method. They do not evaluate human reader performance with or without AI assistance. The device is intended for non-laboratory personnel to use for direct numerical readings.

6. Standalone Performance Study (Algorithm Only Without Human-in-the-Loop Performance)

• Yes, a standalone performance study was done. All the analytical performance studies (precision, linearity, detection limits, analytical specificity, anticoagulant/sample type equivalence, and method comparison with predicate/ICSH) represent standalone performance of the HemoCue® Hb 801 System. The "system" includes the analyzer which uses an algorithm for translation into hemoglobin concentration. Users are described as "non-laboratory personnel" who obtain a direct reading from the device.

7. Type of Ground Truth Used

• The ground truth varied depending on the study:
  • Precision, Linearity, Detection Limits, Analytical Specificity: These studies establish the inherent performance characteristics of the device itself. The "ground truth" for the samples was their known or carefully prepared hemoglobin concentrations as measured by highly controlled laboratory methods.
  • Anticoagulant and Capillary vs. Venous Equivalence: The "ground truth" was the comparison between the two sample types from the same subject.
  • Method Comparison: The ground truth was established by:
    • The predicate device (HemoCue® Hb 301 System).
    • The hemiglobincyanide (HiCN) method, which is the International Council for Standardization in Haematology (ICSH) international reference method.
  • Reference Range Verification: Comparison against published reference intervals from authoritative texts (Dacie and Lewis Practical Haematology, Pediatric Reference Intervals).

8. Sample Size for the Training Set

• Not applicable/Not mentioned. This is an IVD device that measures a specific analyte using a spectrophotometric measuring principle with a pre-defined algorithm and factory calibration. There is no mention of a "training set" in the context of machine learning or AI algorithm development as typically understood in the context of image analysis or diagnostic prediction. The algorithm for translating light measurements into hemoglobin concentration is fundamental to the device's design, not something that appears to be "trained" on a large dataset in the sense of modern AI.

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

• Not applicable/Not mentioned. As noted above, typical "training sets" and their associated ground truth establishment methods (e.g., expert consensus labeling) are not relevant to the description of this device's development or validation. The device's fundamental measurement principle and algorithm are traceable to the HiCN method (ICSH), meaning its design and underlying calculations are based on established scientific principles for hemoglobin determination.

Ask a specific question about this device

The hemochroma PLUS System is for the quantitative determination of hemoglobin concentration in non-anticoagulated capillary (finger-stick) whole blood or venous whole blood (K2-EDTA, sodium citrate, lithium heparin, or sodium heparin). The testing system is designed for point-of-care settings, hospitals, and medical lab facilities.

Estimation of hematocrit, as a function, is only for normal hemoglobin values, 12.0 to 180 g/dL) and in patients ≥ 6 months old.

The hemochroma PLUS Controls are intended for use as quality control material to assure the validity and performance of the hemochroma PLUS system in measuring the human hemoglobin concentration.

The hemochroma PLUS Microcuvettes are only used with hemochroma PLUS Analyzer. The hemochroma PLUS System is for in vitro diagnostic only.

The hemochroma PLUS Analyzer calculates the test result automatically and displays hemoglobin concentration in terms of g/dL.

The hemochroma PLUS Analyzer is a battery powered, hand-held device to measure the concentration of total hemoglobin in blood in 3 seconds with 15uL of whole blood. Whole blood may be collected by fingerstick (capillary) or venipuncture and analyzed without preprocessing. The hemochroma PLUS Analyzer uses hemochroma PLUS Microcuvettes with dual ports where the user applies samples either through capillary action or direct volume pipetting.

The hemochroma PLUS Analyzer determines hemoglobin concentration in whole blood samples using a dual wavelength photo-absorption method and measures the degree of light absorption with a spectrophotometer. The optical distance between the hemochroma PLUS 3 Microcuvette walls is fixed and permits photometric determination of hemoglobin in undiluted blood samples. The computed end result is displayed on the LCD display and can be printed on an external printer (optional).

The hemochroma PLUS System consists of a hemochroma PLUS Analyzer, single-use hemochroma PLUS Microcuvettes, hemochroma PLUS ID Chip, optical System Check Microcuvette and hemochroma PLUS Controls.

Here's a breakdown of the acceptance criteria and the study details for the hemochroma PLUS System, based on the provided document:

Acceptance Criteria and Reported Device Performance

The acceptance criteria are generally established by meeting specific performance metrics determined by the manufacturer, often guided by CLSI (Clinical and Laboratory Standards Institute) guidelines, to ensure accuracy, precision, and reliability. The document details analytical performance studies. The results from the repeatability, reproducibility, linearity, detection limits, and method comparison studies demonstrate that the device meets the defined acceptance criteria, often by being "within the defined acceptance criteria" or showing "comparable performance."

• Within Run: SD (0.09-0.11), %CV (0.47-1.68) for Hgb concentrations 5.6-23.7 g/dL.
• Total: SD (0.20-0.25), %CV (1.06-3.67) for Hgb concentrations 5.6-23.7 g/dL.
  Reproducibility (Across sites, operators, and days):
• Site 1: Total %CV (1.08-1.99) for Hgb controls 8.5-15.8 g/dL.
• Site 2: Total %CV (1.08-1.74) for Hgb controls 8.5-15.8 g/dL.
• Site 3: Total %CV (1.14-2.10) for Hgb controls 8.5-15.8 g/dL.
• Combined Sites: Total %CV (1.23-2.30) for Hgb controls 8.4-15.8 g/dL. Performance results were "within the defined acceptance criteria." |
  | Linearity/Assay Range | Linearity across the claimed measuring range | Linear regression demonstrating acceptable correlation. | Linearity: Linear regression performed on eleven hemoglobin concentration levels (2.5-25.6 g/dL) demonstrated linearity over the claimed measuring range of 5.0-25.6 g/dL. |
  | Detection Limits | Limit of Blank (LoB) | Explicit acceptance criteria not given, but a calculated value is provided. | LoB: 0.23 g/dL |
  | | Limit of Detection (LoD) | Explicit acceptance criteria not given, but a calculated value is provided. | LoD: 1.66 g/dL |
  | | Limit of Quantitation (LoQ) | % Total-error smaller than the desired total error for the measurand. | LoQ: 4.5 g/dL (data considered acceptable as % Total-error was smaller than desired total error). |
  | Analytical Specificity | Interference by exogenous and endogenous substances | Non-significant interference up to specified concentrations. | Interference Study: All tested interference substances (endogenous and exogenous) showed non-significant interference up to the specified concentrations. |
  | Method Comparison | Agreement with predicate device (HemoCue Hb 301 System) | Linear regression demonstrating comparable performance (implied acceptance within a certain slope, intercept, and R-value). | Method Comparison: Linear regression analyses showed comparable performance. Example (Site 1 Capillary): Slope = 0.9942 (95% CI: 0.941-1.048), Intercept = 0.1214 (95% CI: -0.650-0.892), r = 0.980. The study demonstrated that analytical performance is "substantially equivalent" to the predicate device. |
  | Matrix Comparison | Comparability between venous and capillary whole blood samples | Results of Bland-Altman plot analysis and % difference meeting acceptance criteria. | Matrix Comparison: Results of Bland-Altman plot analysis and % difference between venous whole blood samples and capillary whole blood samples met the acceptance criteria. |
  | Sample Stability | Stability of blood samples stored at 2-8°C | Not explicitly stated, but based on recovery. | Sample Stability: Supports a stability claim of 24 hours when stored at 2-8°C. |
  | Anticoagulant Comparison | Agreement between K2EDTA and other anticoagulants | Results of Bland-Altman plot analysis and % difference meeting acceptance criteria. | Anticoagulant Comparison: Results of Bland-Altman plot analysis and % difference between K2EDTA and 4 other anticoagulant tubes were "within the defined acceptance criteria." |

Study Details:

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

   • Repeatability: 5 test samples (ranging from 5.6 g/dL to 23.7 g/dL) were tested 84 times each (total of 420 measurements per study). Data Provenance: In-house (presumably Republic of Korea, where the sponsor is located) and retrospective (prepared samples).
   • Reproducibility: 3 control levels (low, middle, high) were tested with 160 results per control level per site (total 480 results per control level across all 3 sites). Data Provenance: Three point-of-care clinical sites in the United States. Prospective.
   • Linearity/Assay Reportable Range: 11 hemoglobin concentration levels tested in triplicate. Data Provenance: Not explicitly stated, but in-house testing. Retrospective.
   • Detection Limit (LoB, LoD, LoQ):
     • LoB: 5 blank samples, 5 replicates, 3 days, 3 microcuvette lots, 3 analyzers (total 75 results per microcuvette lot).
     • LoD: 6 Hgb-low samples, 5 replicates, 3 days, 3 microcuvette lots, 1 analyzer (total 90 results per microcuvette lot).
     • LoQ: 6 low Hgb samples, 5 replicates, 3 days, 3 microcuvette lots, 1 analyzer (total 90 results per microcuvette lot).
       Data Provenance: Not explicitly stated, but in-house testing. Retrospective/prepared samples.
   • Analytical Specificity (Interference): 3 hemoglobin levels of human whole blood samples, tested in 5 replicates. Data Provenance: Not explicitly stated, but in-house testing. Retrospective/prepared samples.
   • Method Comparison: 60 capillary finger-stick blood samples and 70 venous blood samples (including 10 spiked extreme range samples). Data Provenance: Three point-of-care clinical sites in the United States. Prospective.
   • Matrix Comparison: 80 study participants (venous and capillary whole blood). Data Provenance: Not explicitly stated, likely clinical sites in the United States. Prospective.
   • Sample Stability: 37 fresh venous blood samples. Data Provenance: Not explicitly stated, but in-house testing. Prospective.
   • Anticoagulant Comparison: Venous blood collected from 50 study participants. Data Provenance: Not explicitly stated. Prospective.
   • Disease Conditions Comparison: 3 specimens from Polycythemia, 2 from hypochromia, 3 from high WBC count, 2 from sickle cell donors. Each tested 5 times. Data Provenance: Not explicitly stated. Retrospective.

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

   • For most analytical performance studies (precision, linearity, detection limits, interference), the "ground truth" is typically established by the carefully prepared samples/controls according to standardized procedures (e.g., using reference materials or precise spiking methods) rather than expert consensus on individual case interpretation.
   • For Method Comparison and Anticoagulant Comparison, the predicate device (HemoCue Hb 301 System) serves as the reference ("ground truth") for comparison. The document does not mention the use of human experts to establish ground truth for individual cases, but rather relies on the established accuracy of the predicate device.
   • For Reproducibility at clinical sites, data was collected by "three operators (one at each site)," but their qualifications are not specified beyond being operators.

3. Adjudication Method for the Test Set:

   • No adjudication method (like 2+1 or 3+1 consensus) is described, as the studies primarily involve quantitative measurements and comparison to a reference method (predicate device) or internally established values for controls/calibrators, rather than subjective interpretations by multiple experts.

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

   • No. The studies described are primarily analytical performance studies comparing the device's measurements to a reference method (the predicate device) or established laboratory standards. There is no mention of a human-in-the-loop study assessing improved reader performance with or without AI assistance. This device is an automated hemoglobin analysis system, not an AI interpretation tool for imaging or other diagnostic data that typically involves human readers.

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

   • Yes, all the described analytical and clinical performance studies (precision, linearity, detection limits, interference, method comparison, matrix comparison, stability studies, anticoagulant comparison, disease conditions comparison) are conducted to assess the performance of the device itself (algorithm + hardware) in a standalone manner, without explicit human interpretive involvement in the result generation or assessment beyond operating the device.

6. The Type of Ground Truth Used:

   • Reference Method: For method comparison, the HemoCue Hb 301 System was used as the reference method.
   • Prepared Samples/Controls: For precision, linearity, detection limits, and interference studies, ground truth was implicitly established through the careful preparation of samples with known hemoglobin concentrations or the use of quality control materials with assigned values.
   • Natural Samples: Many studies utilized "natural human whole blood samples" or "fresh venous blood samples," for which the "ground truth" would be the measured value by the hemochroma PLUS in initial readings, or by the predicate device for comparative studies.

7. The Sample Size for the Training Set:

   • The document primarily describes validation studies for a device, not the development of an AI algorithm which requires a separate training set. The device itself uses a "pre-programmed calibration" and an "ID chip" with "calibration data/information." The "Value Assignment" section for the hemochroma PLUS Controls used 15 replicates for each control level to set the mean values, and then 10 replicates on each of three analyzers with three microcuvette lots to verify these values. This isn't a "training set" in the context of machine learning, but rather establishing performance characteristics for physical controls used with a calibrated instrument.

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

   • As noted above, this device is not an AI/machine learning system that requires a "training set" in the conventional sense. Its "ground truth" for calibration and control value assignment is established through standard laboratory practices, including testing in multiple replicates, using multiple lots of reagents/devices, and setting mean values through statistical analysis. The device uses "pre-programmed calibration" and calibration data from its ID chip, which would have been established by the manufacturer using reference methods and standard calibration procedures.

Ask a specific question about this device