Search Results

Not Found

Ventana Anti-Keratin-AE1 Primary Antibody (Clone AE1) for use on the Ventana ES Automated Slide Stainer.

Here's an analysis of the provided text regarding the acceptance criteria and study for the Ventana Anti-Keratin-AE1 Primary Antibody (Clone AE1):

Based on the provided 510(k) summary, the acceptance criteria and the study conducted are quite different from what would typically be expected for an AI device. This document describes a primary antibody used in immunohistochemistry, not an AI-powered diagnostic device. Therefore, many of the requested categories related to AI performance, ground truth establishment for AI, and expert involvement/adjudication for AI will not be directly applicable or present in this summary.

I will populate the table and answer the questions based on the information provided in the text, highlighting where the information is not relevant to an AI device or is missing.

Acceptance Criteria and Reported Device Performance

Given this is a primary antibody, the "acceptance criteria" are implied by the comparative study's findings regarding specificity, sensitivity, and reproducibility when compared to a predicate device.

Study Details

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

   • Test Set Sample Size:
     • Pathologic Samples: Tested for staining included breast carcinomas (at least 18), melanomas, squamous cell carcinomas, carcinoids, and leiomyosarcomas. The text specifically states "18 of 18 epithelial cancers" for Anti-Keratin-AE1.
     • Normal Samples: Breast, ureter, thyroid, skin, small intestine, stomach, liver, smooth muscle, prostate, tonsil, pituitary, thymus, esophagus, ovary, testes, pancreas, cardiac muscle, spinal cord, spleen, adenoid, and kidney (number of individual samples for each not specified).
     • Reproducibility Studies: 15 samples for inter-run and 10 samples for intra-run reproducibility, using "the same tissue".
   • Data Provenance: "Samples were obtained from excess tissues obtained for reasons other than the present study." This indicates a retrospective collection of samples. The country of origin is not specified but is presumably the US given the 510(k) submission.

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

   • The document implies that an expert (e.g., a pathologist) evaluated the slides for "specific staining intensity and background staining." However, the number of experts is not specified, nor are their qualifications. For an antibody, the "ground truth" (i.e., whether a cell is epithelial and should stain) is based on standard histological and pathological assessment, which is performed by qualified clinicians, but the study does not detail this aspect of setup.

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

   • None specified. The text does not describe any formal adjudication method for the evaluation of the staining results.

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 was not an MRMC comparative effectiveness study involving human readers with and without AI assistance. This study is for a primary antibody, not an AI device.

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

   • Not applicable. This device is a primary antibody, not an algorithm. Therefore, "standalone algorithm performance" is not a relevant concept for this product. Its performance is evaluated by its chemical/biological interaction with tissue.

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

   • The "ground truth" for the tissues tested would be pathology-based diagnosis (e.g., "this is a breast carcinoma," "this is normal kidney tissue"). The evaluation of staining intensity and specificity would be based on expert visual assessment against established pathological criteria.

7. The sample size for the training set:

   • Not applicable. This is an antibody, not a machine learning model. There is no concept of a "training set" for this type of medical device. The "training" for such a product would be its development and formulation.

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

   • Not applicable. As there is no training set for an AI model, the question of establishing ground truth for it is irrelevant in this context.

Ask a specific question about this device