Search Results

"OC-Auto Sensor io iFOB Test" is designed to be used together as an immunoassay test system. The test system is intended for the qualitatitye detection of fecal occult blood in feces by professional laboratories. The automated test is used for the measurement of fecal occult blood and is useful as an aid to detect blood in stool when lower gastrointestinal bleeding may be suspected.

OC-Auto Sensor io iFOB Test is intended for the automated in vitro qualitative detection of fecal occult blood in feces by professional laboratories. The test system is comprised of test reagents (latex, diluent buffer, wash concentrate, calibrator, negative and positive controls), sample collection bottles and analyzer.

The principle of measurement employed for the reagent system is latex agglutination. A latex agglutination reaction is the clumping of antibody-sensitized polystyrene latex particles through an antigen-antibody reaction. A light beam is passed through the reaction liquid to measure changes in the intensity of the transmitted light beam (latex turbidimetry), and changes in the intensity of the scattered light beam (latex nephelometry). With OC-Auto Sensor io iFOB Test analyzer, latex turbidimetry is used to measure the amount of an antigen or an antibody by measuring changes in scattered light rays in latex agglutination.

The throughput of the instrument is 88 samples per hour. The samples are collected in the sample collection bottles that are sent home with the patient. The sample collection bottles are then returned to the laboratory. The inverted sample collection bottles are racked and placed onto the instrument platform. The sample collection bottle is punctured and a sample is pipet into the cuvette followed by the latex reagent and mixed. Measurements are taken between the mixing cycles. After a series of washes the blank is read and the final results calculated and printed.

Here's an analysis of the acceptance criteria and the study that proves the device meets them, based on the provided text:

Acceptance Criteria and Device Performance

The acceptance criteria for the OC-Auto Sensor io iFOB Test are largely demonstrated through its equivalence to the predicate device and robust validation of its performance characteristics. The specific acceptance criteria themselves are not explicitly listed in a table format with quantitative targets for each category. However, the performance characteristics studies confirm that the device meets implied acceptance levels by showing high agreement with expected values and the predicate device.

Table of Acceptance Criteria and Reported Device Performance:

• Negative Samples (0, 50, 80 ng/mL): 100.0% Negative Percentage Agreement (99.5% ~ 100.0% CI for individual sites, 99.9% ~ 100.0% CI for all clinical sites combined).
• Positive Samples (120, 450, 700 ng/mL): Very high Positive Percentage Agreement.
  • 120 ng/mL: 99.8% - 100.0% PPA (99.1% ~ 99.9% CI to 99.5% ~ 100.0% CI for individual sites, 99.7% ~ 100.0% CI for all clinical sites combined).
  • 450 ng/mL: 100.0% PPA (99.5% ~ 100.0% CI for individual sites, 99.9% ~ 100.0% CI for all clinical sites combined).
  • 700 ng/mL: 100.0% PPA (99.5% ~ 100.0% CI for individual sites, 99.9% ~ 100.0% CI for all clinical sites combined).
• The negative/positive threshold (100 ng/mL) showed variable distribution of positive/negative results, which is expected at the cutoff. Overall Percentage Agreement for this range was 100.0% (99.9% ~ 100.0% CI).
• All test results satisfied the acceptance criteria (stated in the text). |
  | Linearity | Measured values should align with theoretical values across the reported detection range. | Measured values were treated as regression values and compared against theoretical values (intended hemoglobin concentration from dilution). The test results satisfied the criteria. (Specific quantitative results not provided, but deemed acceptable). |
  | Prozone Effect | No susceptibility to prozone effect within the specified concentration range. | Device is not susceptible to prozone effect up to 1953 ng/mL. |
  | Limit of Detection (LoD) | Ability to detect hemoglobin at a specific low concentration. | 20 ng/mL was determined as the limit of detection. |
  | Hemoglobin Variants | Equivalence in sensitivity to common hemoglobin variants (S, C, F). | Device is equally sensitive to hemoglobin S, C, and F. |
  | Cross Reactivity | No false positives or interference from animal hemoglobin, vegetable extracts, or meat extracts. | Animal Hemoglobin: No cross reactivity with bovine, equine, porcine, goat, sheep, rabbit, turkey, and fish hemoglobin.
  Vegetable Extracts: No cross reactivity with broccoli, cauliflower, cantaloupe, horseradish, red radish, parsnip, and turnip extracts.
  Animal Meat Extracts: No interference with beef, pork, chicken, lamb, and fish extracts. |
  | Interference | No interference from common toilet cleaners, drugs, and dietary supplements. | Toilet Cleaners: No interference with Nuriper, Lysol Bleach, and Blue Enzyme.
  Drugs and Dietary Supplements: No interference with Iron, Vitamin C, laxative, glycerol concentration for enema, and peroxidase. |
  | Stability (Reagents) | Reagents maintain performance over their labeled shelf life. | Stable for 12 months at 2-8°C (Latex Reagent and Buffer). |
  | Stability (Calibrator) | Calibrator maintains performance over its labeled shelf life. | Stable for 12 months at 2-8°C. |
  | Stability (Controls) | Controls maintain performance over their labeled shelf life. | Stable for 12 months at 2-8°C (Positive and Negative Controls). |
  | Stability (Sampling Bottle) | Sampling bottle maintains integrity and sample stability over its labeled shelf life and under simulated extreme shipping conditions. | Stable for 18 months at 2-30°C.
  Inoculated Native Sample Stability: Samples stable for 15 days at room temperature, and 30 days when refrigerated.
  Inoculated Sample Shipping Test: Samples stable for 15 days during shipment under simulated extreme heat conditions. |
  | Reagent Open Bottle Stability | Reagents maintain performance after opening for a specified period on-board the analyzer. | Stable for 7 days after opening and kept on-board. |
  | Humidity Effect | No adverse effect of humidity on reagent stability. | No effect of humidity (25%, 50%, 80% at 23°C) on reagents (latex, buffer, calibrator, controls). |
  | Method Comparison | Substantial equivalence to the predicate device in terms of diagnostic performance (PPA, NPA, OPA). | Overall percent agreement (OPA) was 100 % (95 % CI 99.1 - 100 %), with positive percent agreement (PPA) 100 % (95 % CI 96.9 - 100 %), and negative percent agreement (NPA) 100 % (95 % CI 98.8 - 100 %). For CRC patients' samples, PPA was 100% (95% CI 79.6% - 100%) and NPA was 100% (95% CI 56.6% - 100%), with OPA 100% (95% CI 83.9% - 100%). The study demonstrated that the analytical performance of the device is substantially equivalent to the predicate. |
  | Cybersecurity | Immune to cyberattacks via network, secure USB/RS-232C terminals. | No network connecting function. USB memory and RS-232C terminals are for data output only and have no read functions. |
  | Electromagnetic Compatibility (EMC) | Meets relevant EMC standards. | Passed tests at Power Frequency Magnetic Field 30 A/m and Electrostatic Discharge ±2 kV, ±4 kV, ±8 kV contact; ±2 kV, ±8 kV, ±15 kV air. |

Study Information:

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

• Precision/Reproducibility:
  • For each of the seven known concentrations (0, 50, 80, 100, 120, 450, 700 ng/mL): 21 replicates were measured.
  • This was performed daily (morning and afternoon) over 20 days.
  • Total individual measurements per concentration per site = 21 (replicates) * 2 (times/day) * 20 (days) = 840 measurements.
  • Total measurements across 3 clinical sites for each concentration = 2520 measurements.
  • Data Provenance: Not explicitly stated, but performed "in-house and in three intended use sites." The "intended use sites" typically imply clinical laboratories, likely in Japan (country of origin of manufacturer) or the US (for regulatory submission). The study is prospective in nature as samples were intentionally prepared and tested.
• Linearity, Prozone Effect, Limit of Detection, Hemoglobin Variants, Cross Reactivity, Interference:
  • For each specific condition/concentration tested: 21 replicates were measured.
  • Data Provenance: In-house studies. Prospective.
• Stability Studies (Reagents, Calibrator, Controls, Open Bottle):
  • For each time point and condition: 21 replicates of stool samples spiked with the seven known hemoglobin concentrations were measured.
  • Data Provenance: In-house studies. Prospective.
• Inoculated Native Sample Stability:
  • Not explicitly stated, but samples were prepared with 6 hemoglobin concentrations similar to the 7 known concentrations (excluding 0, presumably, or similar). Replicates are implied.
  • Data Provenance: In-house studies. Prospective.
• Inoculated Sample Shipping Test:
  • Samples prepared in the same way as the native sample stability study (implying similar replicates/conditions).
  • Data Provenance: In-house simulated study. Prospective.
• Method Comparison:
  • Total samples: 425 samples.
  • This included 20 CRC patients samples.
  • Data Provenance: Performed at "one professional medical laboratory in the U.S. and two international professional medical laboratories." This indicates prospective collection of samples used for the comparison study, though the samples themselves might have been collected retrospectively from patients or prepared for the study.

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

• For the Precision/Reproducibility, Linearity, Prozone Effect, Limit of Detection, Hemoglobin Variants, Cross Reactivity, Interference, Stability studies, and Inoculated Native Sample Stability/Shipping tests, the ground truth was established by preparation of controlled samples with known concentrations of hemoglobin or interfering substances. No human expert consensus was required for these analytical performance studies.
• For the Method Comparison study, the ground truth was the predicate device's result. The predicate device (OC-Sensor DIANA iFOB Tes, K092330) itself would have been validated against a clinical ground truth (e.g., colonoscopy, pathology) in its own clearance process, but for this specific study, the predicate served as the reference standard. Thus, no new experts were used to establish ground truth in this comparative effectiveness study.

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

• No explicit adjudication method (like 2+1, 3+1) is mentioned. This is typical for in vitro diagnostic (IVD) devices where results are quantitative or qualitative (positive/negative) based on pre-defined cutoffs, rather than subjective interpretations by multiple readers.
• For the precision studies, the "expected value" (negative/positive) served as the reference for agreement. For the method comparison, the predicate device result served as the reference.

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 MRMC comparative effectiveness study was done.
• This device is an automated in vitro diagnostic (IVD) test system. It performs qualitative detection of fecal occult blood using immunoassay, meaning the results are determined by the analyzer itself, not through human interpretation of images or complex data that AI would assist with. The "AI" would be the automated algorithm within the device for analysis, and its performance is evaluated as a standalone system.

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

• Yes, the entire set of performance studies (Precision, Linearity, Prozone Effect, LoD, Stability, Cross-reactivity, Interference) represents a standalone performance evaluation. The device (OC-Auto Sensor io iFOB Test system, including the analyzer and reagents) performs the analysis and provides results automatically without human interpretation in the decision-making loop for individual sample outcomes. The method comparison also evaluates the standalone performance against a predicate standalone device.

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

• For most analytical performance studies (Precision, LoD, etc.): The ground truth was controlled samples with known, spiked concentrations of human hemoglobin or interfering substances.
• For the method comparison study: The ground truth was the results obtained from the predicate device (OC-Sensor DIANA iFOB Tes). While the predicate's original clearance would have relied on clinical correlation (potentially pathology or outcomes data), for this specific 510(k) submission, the predicate itself served as the reference.

7. The sample size for the training set:

• The document describes performance studies and comparisons, but does not explicitly mention a "training set" or "validation set" in the context of machine learning. This is because the device is an immunoassay system, not an AI/ML-based diagnostic software. Its underlying principles are based on known chemical reactions and optical density measurements, which are "trained" through calibration curves rather than an algorithm trained on large datasets.
• The "calibration curve" is established using 5 points: 0, 50, 200, 500, 1000 ng Hb/mL.

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

• As above, there isn't a traditional "training set" as understood in AI/ML. The device is calibrated.
• The ground truth for calibration is established using purified hemoglobin in buffer at precisely known concentrations (0, 50, 200, 500, 1000 ng Hb/mL). The 1000 ng/mL calibrator is derived from human blood and tested to be negative for HBS antigens, HIV, and HCV antibodies.

Ask a specific question about this device