(147 days)
Intelligent Medical Imaging, Inc.'s MICRO21™ with Cerebrospinal Fluid (CSF) is a laboratory instrument used to perform cytological analysis by locating, digitally storing and displaying nucleated cells in human cerebrospinal fluid preparations. Examination of cellular morphology and classification must be performed by qualified individuals.
The MICRO21™ with Cerebrospinal Fluid is a new procedure for the MICRUZ/ with the tollowing in vitro diagnostic intended use: Intelligent Medical Imaging, Inc.'s MICRO21™ with Cerebrospinal Fluid (CSF) is a laboratory instrument used to perform cytological analysis by locating, digitally storing and displaying nucleated cells in human cerebrospinal fluid preparations. Examination of cellular morphology and classification must be performed by qualified individuals.
A summary of the MICRO21 with Cerebrospinal Fluid process is as follows:
- Patient CSF samples are prepared using the cytocentrifugation method for specimen concentration 1. using the Cytospin® centrifuge. . .
- Each slide has a black mask over the non-frosted portion with a hole in the mask for specimen 2. placement.
-
- The slide is stained using Wright's stain.
- The slides are barcoded and placed into a frame holder, and the frame is placed on the MICRO21 for 4. processing.
- ડ. The MICRO21 locates nucleated cells.
- The nucleated cell images are stored by the instrument and displayed on a color monitor for review by 6. a technologist.
- The technologist reviews the images and confirms each cell type. 7.
-
- A report of the CSF result for each patient is printed.
The provided text describes information relevant to the regulatory submission for the MICRO21™ with Cerebrospinal Fluid, an automated cell locating device. Based on the document, here's a structured breakdown of the acceptance criteria and the study conducted:
1. Table of Acceptance Criteria and Reported Device Performance
The document does not explicitly present a table of quantitative acceptance criteria with corresponding performance metrics. Instead, it describes various test methods and concludes that these tests "confirm the safety and effectiveness" for the intended use. Therefore, the "acceptance criteria" are implied by the successful completion and correlation reported in the "Conclusion" section for the specified tests.
| Test Method | Implied Acceptance Criteria (Qualitative) | Reported Device Performance |
|---|---|---|
| Method Verification and Sensitivity | Ability to locate cells at low and high power, distinguishing real cells from non-cells (e.g., stain precipitate). | "The ability of the instrument to locate cells at low and high power was determined by manually confirming the location of real cells or the location of non-cells such as stain precipitate." (Implies successful confirmation of ability to locate and differentiate). |
| Precision | Consistent and reproducible results (intra-instrument within slide precision) on repeated analyses of the same specimens by the device. | "The results were correlated to provide intra-instrument within slide precision." (Implies satisfactory precision, as it supports the overall conclusion of safety and effectiveness). |
| Correlation Study Performance | Satisfactory correlation between Cerebrospinal Cell Differentials (CCDs) obtained from MICRO21 images and those performed manually by qualified technologists. | "CCDs obtained from images located and displayed by the MICRO21 were correlated with CCDs performed manually by qualified technologists." (Implies successful correlation, as it supports the overall conclusion of safety and effectiveness. No specific correlation coefficient is given). |
2. Sample Size and Data Provenance for the Test Set
-
Test Method 1 (Method Verification and Sensitivity):
- Sample Size: "Ten slides" were processed.
- Data Provenance: Not explicitly stated, but implied to be human CSF preparations for testing the device's main function. Retrospective or prospective is not specified.
-
Test Method 2 (Precision):
- Sample Size: "Three CSF specimen slides" were processed.
- Data Provenance: Not explicitly stated. Retrospective or prospective is not specified.
-
Test Method 3 (Correlation Study Performance):
- Sample Size: "Fifty CSF specimens" were used.
- Data Provenance: "obtained from various geographic sites." This suggests a multi-site data collection, which could be a mix of retrospective or prospective data. The document does not specify.
3. Number of Experts and Qualifications for Ground Truth (Test Set)
- Test Method 1 (Method Verification and Sensitivity): The manual confirmation was done by an unspecified number of individuals, but implies medical professionals capable of identifying cells and precipitates. No specific qualifications are given.
- Test Method 2 (Precision): "The resulting images were presented to a technologist for review." Implies one technologist. Qualifications are "technologist," with no further detail.
- Test Method 3 (Correlation Study Performance):
- Manual Method Ground Truth: "two technologists processed and examined each slide by bright-field microscopy." Qualifications: "qualified technologists." (No specific experience level or certifications provided beyond "qualified").
- MICRO21 Image Review: "two technologists independently reviewed the stored images for each specimen and classified all nucleated cells accordingly." Qualifications: "technologists."
4. Adjudication Method (Test Set)
- Test Method 1 (Method Verification and Sensitivity): "manually confirming the location of real cells or the location of non-cells". The adjudication method is not specified (e.g., if there were disagreements, how they were resolved).
- Test Method 2 (Precision): Not applicable for explicit adjudication as it focuses on instrument reproducibility, with a single technologist reviewing results.
- Test Method 3 (Correlation Study Performance):
- Manual Method: "two technologists processed and examined each slide". It's not explicitly stated if there was an adjudication process if their classifications differed. It is possible they performed their counts and classifications independently, and the aggregated results were used for correlation.
- MICRO21 Image Review: "two technologists independently reviewed the stored images... and classified all nucleated cells accordingly." Similar to the manual method, it's not specified if there was an adjudication process for discrepancies between the two technologists' independent reviews.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No, a multi-reader multi-case (MRMC) comparative effectiveness study as typically understood (comparing human readers with AI assistance vs. without AI assistance to measure an effect size) was not explicitly described.
The "Correlation Study Performance" compared the device's output (images reviewed by technologists) with the manual method performed by technologists. While it involves multiple readers (technologists) and multiple cases (50 CSF specimens), it's more of a standalone performance evaluation of the device in a clinical workflow context, aiming to show agreement with current standard practice, rather than an MRMC study designed to quantify the improvement of human readers through AI assistance. The device's role is to locate, store, and display cells, with technologists performing the final classification.
6. Standalone Performance Study
Yes, a standalone (algorithm only without human-in-the-loop performance) study was performed for the locating component of the device.
- Test Method 1 (Method Verification and Sensitivity): This test determined "the ability of the instrument to locate cells at low and high power." While manual confirmation was used to verify the accuracy of the locations, the initial "locating" was an algorithm-only step.
- Test Method 2 (Precision): This test assessed the instrument's intra-instrument precision, focusing on the reproducibility of the locating and imaging process, which is largely an algorithmic function.
Even for the "Correlation Study Performance," the MICRO21 "located" and "captured" the images. The subsequent human technologists reviewed these captured images. Thus, the locating capability itself is a standalone function of the algorithm, which was then presented to human readers for final classification.
7. Type of Ground Truth Used
The ground truth for the test set was:
- Expert Consensus/Expert Review: For all test methods, the ultimate "ground truth" or reference standard involved human experts (technologists) making classifications or confirming locations.
- For Test Method 1 and 2, it involved technologists manually confirming cell locations or reviewing images.
- For Test Method 3 (Correlation Study), the "manual method" performed by two qualified technologists served as the reference standard against which the device's performance (images reviewed by other technologists) was compared. While "consensus" isn't explicitly stated for the manual method, it's implied that their independent processing and examination established the manual CCDs.
8. Sample Size for the Training Set
The document does not provide information about the sample size used for the training set of the MICRO21™ device. This information is typically found in development documentation and is often distinct from the regulatory submission's test set.
9. How 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. Given the device's function (locating cells and displaying images for human review), it's highly probable that the training data would have involved slides prepared with known cellular content, manually annotated or classified by expert cytotechnologists or pathologists to teach the system to identify and locate target cells. However, this is an inference and not explicitly stated in the provided text.
§ 864.5260 Automated cell-locating device.
(a)
Identification. An automated cell-locating device is a device used to locate blood cells on a peripheral blood smear, allowing the operator to identify and classify each cell according to type. (Peripheral blood is blood circulating in one of the body's extremities, such as the arm.)(b)
Classification. Class II (performance standards).