Search Results

Automated scanning microscopy work station for histological identification of the enzyme leucocyte alkaline phosphatase in neutrophilic granulocytes to differentiate: Granulocytic leukemia; a malignant disease characterized by excessive overgrowth of granulocytes in bone marrow, and Reactions that resemble true leukemia, such as those occurring in severe infections.

The MD, 2000 Digital Analyzer is an automated intelligent microscope cell locating device that detects, by color and pattern recognition techniques, cells stained with alkaline phosphatase reagent system FAST VIOLET B SALT. The system consists of software resident in computer memory and includes keyboard, color monitors, microscope, printer, and an automatic slide handling and scanning mechanism controlled and operated by a health care professional.

The provided document describes the MDx 2000 Digital Analyzer, an automated intelligent microscope cell locating device. The document details the device's intended use and the clinical trials conducted to demonstrate its substantial equivalence to predicate devices. However, it does not explicitly state quantitative "acceptance criteria" for performance that the device was designed to meet. Instead, the study aims to show that the MDx 2000 performs "as well or better than the predicate devices."

Here's an analysis of the available information:

1. Table of Acceptance Criteria and Reported Device Performance

As mentioned, explicit, quantitative acceptance criteria are not presented in this summary. The evaluation focuses on demonstrating comparable performance and improvements over manual methods rather than meeting predefined numerical thresholds.

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

• Test Set Sample Size: 125 patients tested across 3 clinical sites.
• Data Provenance: Clinical trials were conducted at three different sites. There is no explicit mention of the country of origin, but given the context of a 510(k) summary for the U.S. FDA, it is highly likely the data is from the United States. The study was prospective in nature, as indicated by the "written MicroVision Medical Systems, Inc. protocol" and controlled clinical trials.

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

The document mentions "qualified medical technologists examining the slides" for the manual comparison (The number of qualified medical technologists examining the slides). However, it does not specify the exact number of experts or their specific qualifications (e.g., years of experience) used to establish the ground truth for the test set.

4. Adjudication Method for the Test Set

The document states, "The design of the double blind study." While a double-blind study suggests measures to reduce bias, it does not explicitly describe an adjudication method (like 2+1 or 3+1). The "manual count" by qualified technologists serves as the reference, but the process for resolving discrepancies among manual readers (if multiple were involved in the ground truth) is not detailed.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done, and the Effect Size

The summary does not explicitly describe a Multi-Reader Multi-Case (MRMC) comparative effectiveness study in the typical sense of measuring human reader improvement with AI assistance versus without. The study primarily compares the automated device's performance to a manual method performed by technologists.

However, the findings imply an "effect size" related to variability:

• The device "eliminated the considerable between-technologist variability that occurred at each of the study sites." This indicates a significant positive effect on consistency compared to unassisted human readers.
• The device "provided more between-site consistency in reported results than that for the comparative Sigma NAP kit manual method." This also suggests an improvement in consistency over a purely manual process.
  The document doesn't provide a quantitative effect size (e.g., percentage improvement in accuracy or reduction in reading time) due to AI assistance specifically.

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

Yes, the study primarily evaluates the standalone performance of the MDx 2000 Digital Analyzer algorithm (the "candidate device") against manual methods. The device is described as an "automated intelligent microscope cell locating device" that detects cells using "color and pattern recognition techniques." The assessments of precision, consistency, and accuracy are all of the automated system's output.

7. The Type of Ground Truth Used

The ground truth for the clinical trials was established by:

• Expert Consensus/Manual Count: The "average manual count" performed by "qualified medical technologists" based on "a scoring procedure... from ratings from zero to 4+ on the basis of quantity and intensity of precipitated dye within the cytoplasm of the cells." This manual method was based on established cytochemical techniques and a "scoring procedure" detailed in referenced literature.

8. The Sample Size for the Training Set

The document does not provide information on the sample size used for the training set. The focus of this 510(k) summary is on the clinical validation of the device, not its development or training process.

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

Since the document does not mention the training set size, it also does not specify how the ground truth for the training set was established. We can infer that the device's "color and pattern recognition techniques" would have been developed using some form of labeled data, likely derived from expert analysis of stained cells, but the specifics are not included in this regulatory submission summary.

Ask a specific question about this device