Search Results

The Duet™ System is an automated scanning microscope and image analysis system. It is intended for in-vitro diagnostic usc as an aiding tool to the pathologist in the detection, classification and counting of cells of interest based on color, intensity, size, pattern, and shape. The Duet™ System is intended to detect:

1. Hematopoietic cells stained by Giemsa stain, Immunohistochemistry or ISH (with bright field and fluorescent) prepared from ccll suspension.
2. Amniotic cells stained by FISH (using direct labeled DNA probes for chromosomes X, Y, 13, 18 and 21).
3. Cells in urine specimens, stained by FISH (using the Vysis UroVysion™ Bladder Cancer Recurrence Kit for chromosomes 3, 7, 17 and 9p21 locus), from subjects with transitional cell carcinoma of the bladder.

The Duet™ System is a fully integrated imaging and scanning platform that automates time-consuming and difficult laboratory tasks of slide screening by making a significant reduction in time and labor currently required. The Duet™ scans in high resolution and in full color cell samples at high speed both in bright light illumination and in fluorescent illumination. Duet™ suggests classification of the cells according to their morphological features, their staining (Giemsa, IHC) and fluorescent signals, and allows the user to quickly examine the results, correct them as needed and generate a report summarizing the sample's data. The unique feature of the Duet system allows the presents combined presentation of morphological and specific staining information of the same cell, for all the cells of the sample.

Here's a detailed breakdown of the acceptance criteria and the studies performed for the Duet™ system, based on the provided 510(k) summary:

1. Table of Acceptance Criteria and Reported Device Performance

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

• Test Set Sample Size:
  • Study 1 (Comparison to Manual Scoring): Total of 135 slides, with some slides screened twice, resulting in a total of 178 slides for statistical analysis.
  • Study 2 (Reproducibility & Repeatability): 4 slides (2 negative, 1 positive, 1 near medical decision cutoff).
• Data Provenance: The document doesn't explicitly state the country of origin for all patient data. However, Study 2 was performed at Bio View Lab in cooperation with Meir Hospital, Kfar Saba, Israel, indicating at least some data from Israel. The nature of the studies (comparison to manual methods, reproducibility) suggests these were primarily retrospective analyses of prepared slides.

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

• Study 1 (Comparison to Manual Scoring):
  • The ground truth was established by the "manual method" of scoring.
  • The document states that detection, classification, and counting are "routinely performed manually by cytogeneticist using conventional microscopes, according to the kit instructions." This implies the ground truth was set by one or more cytogeneticists, who are qualified professionals in this field. The exact number of cytogeneticists involved in establishing the ground truth for each slide is not specified (e.g., if it was a single expert per manual read or a consensus).
• Study 2 (Reproducibility & Repeatability): The ground truth for the 4 slides was implied by their selection (2 negative, 1 positive, 1 near medical decision cutoff), presumably based on prior expert evaluation. The experiment was performed by two (2) operators, but their role was in operating the system and performing the analysis, not necessarily establishing the initial ground truth for slide classification.

4. Adjudication Method for the Test Set

• The document does not explicitly describe an adjudication method for reconciling disagreements (e.g., 2+1 or 3+1).
• For Study 1, the "manual method" itself served as the reference standard against which the Duet™ system was compared. Disagreements would have been considered discrepancies between the device and the manual read, not a conflict among multiple ground truth readers requiring adjudication.

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

• The document does not describe an MRMC comparative effectiveness study where human readers' performance with and without AI assistance was evaluated to determine an effect size.
• Study 1 was a comparison of the automated Duet™ system to the "manual scoring method" (human readers performing the task without AI assistance), demonstrating the AI's standalone performance in relation to human performance. It does not measure the improvement of human readers with AI assistance.

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

• Yes, a standalone study was performed (Study 1). The Duet™ system's performance was directly compared to the manual scoring method. The description "Duet™ suggests classification of the cells...and allows the user to quickly examine the results, correct them as needed and generate a report" indicates an assisting tool, but Study 1 assesses the system's ability to classify autonomously before user correction by comparing its output directly to manual scoring. This implicitly evaluates the algorithm's performance.

7. Type of Ground Truth Used

• Expert Consensus (Implicit/Reference Standard): The ground truth for the test sets (especially in Study 1) was established by the "manual method" of cytogeneticists reading slides with conventional microscopes, which represents an expert consensus or standard practice. It is not explicitly stated if multiple experts reached a consensus for each slide used as the "manual method" reference, or if it was based on the standard practice of a single, qualified cytogeneticist.

8. Sample Size for the Training Set

• The document does not specify the sample size used for the training set for the Duet™ system's algorithms. It focuses entirely on the performance evaluation using test sets.

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

• The document does not provide information on how the ground truth for the training set was established. This information is typically proprietary to the developer and not always included in a 510(k) summary focused on validation studies.

Ask a specific question about this device