CELLAVISION DM1200 WITH THE BODY FLUID APPLICTION

K102778 · CellaVision AB · JOY · Sep 16, 2011 · Hematology

Device Facts

• Record ID: K102778
• Device Name: CELLAVISION DM1200 WITH THE BODY FLUID APPLICTION
• Applicant: CellaVision AB
• Product Code: JOY · Hematology
• Decision Date: Sep 16, 2011
• Decision: SESE
• Submission Type: Traditional
• Regulation: 21 CFR 864.5260
• Device Class: Class 2
• Attributes: AI/ML

AI Performance

Output	Acceptance	Observed	Dev DS	Test DS
White Blood Cell Differential Count	Slope between 0.9 – 1.1 and intercept between -0.1 – +0.1 within 95% confidence interval	Neutrophils: y = 0.9969x + 0.0050; Lymphocytes: y = 0.9815x + 0.0016; Eosinophils: y = 1.1048x - 0.0002; Macrophages: y = 1.0067x - 0.0050; Other cells: y = 0.9534 + 0.0032	Not specified	Method comparison study at two clinical sites using 260 samples (62 CSF, 151 Serous fluid, 47 Synovial fluid)
White Blood Cell Differential Count (Repeatability)	0-10%: ≤ 3% SD; 11-60%: ≤ 20% CV; 61-100%: ≤ 8% CV	Results varied by cell class and sample; see precision tables for specific SD/CV values	Not specified	Repeatability study using two CellaVision DM 1200 instruments with 200-cell differential counts repeated 20 times (n=20)
White Blood Cell Differential Count (Reproducibility)	0-10%: ≤ 3% SD; 11-60%: ≤ 20% CV; 61-100%: ≤ 8% CV	All results within acceptance criteria	Not specified	Reproducibility study at three sites using three CellaVision DM 1200 instruments, 12 samples in duplicate, for 20 operating days (n = 120 per sample)

Intended Use

DM1200 is an automated system intended for in-vitro diagnostic use. The body fluid application is intended for differential count of white blood cells. The system automatically locates and presents images of cells on cytocentrifuged body fluid preparations. The operator identifies and verifies the suggested classification of each cell according to type. DM1200 is intended to be used by skilled operators, trained in the use of the device and in recognition of blood cells.

Device Story

Automated system for differential white blood cell (WBC) counting in cytocentrifuged body fluid preparations. Input: stained cell images from slides. Operation: system moves via battlement track pattern to locate/count nucleated cells; uses deterministic artificial neural networks (ANNs) to pre-classify cells into WBC categories (Neutrophils, Eosinophils, Lymphocytes, Macrophages, Other). Output: presented cell images with suggested classifications. Workflow: operator reviews, verifies, accepts, or reclassifies cell images. Used in clinical laboratories by trained personnel. Benefits: standardizes cell identification, improves efficiency of manual differential counts, and provides digital record of cell morphology.

Clinical Evidence

Method comparison study (CLSI EP9-A2) comparing DM1200 (test) to DM96 (reference) using 260 body fluid samples (CSF, serous, synovial). Primary endpoints: accuracy (correlation of cell class counts) and short-term imprecision. Results showed high correlation (R^2 0.9273-0.9932) across all cell classes (Neutrophils, Lymphocytes, Eosinophils, Macrophages, Other). Precision was equivalent between test and reference methods. No clinical diagnostic outcomes reported; bench-based performance validation only.

Technological Characteristics

Automated cell-locating system; optical imaging with 10x/50x zoom; deterministic artificial neural networks (ANN) for cell classification; barcode-labeled slide processing; 12-slide magazine capacity; software-based image storage and database management.

Indications for Use

Indicated for differential white blood cell counting in cytocentrifuged body fluid preparations (CSF, serous, synovial). Intended for use by trained, skilled operators in clinical settings.

Regulatory Classification

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.)

Predicate Devices

• CellaVision® DM96 with the body fluid application (k080595)

Submission Summary (Full Text)

{0}

1

# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY

A. 510(k) Number:
k102778

B. Purpose for Submission:
Modification of device for additional body fluid application

C. Manufacturer and Instrument Name:
CellaVision AB, CellaVision® DM 1200 with the body fluid application

D. Type of Test or Tests Performed:
White Blood Cells differential (peripheral blood smear), RBC characterization, Platelet estimation, White Blood Cells differential (cytocentrifuged body fluid smear).

E. System Descriptions:

1. Device Description:
CellaVision® DM 1200 with the body fluid application is very similar to the predicate (CellaVision® DM96 with the body fluid application, k080595) in terms of technology, functionality, and intended use. The device automatically locates and presents images of nucleated cells on cytocentrifuged body fluid preparations. The system suggests a classification for each cell and the operator verifies the classification and has the opportunity to change the suggested classification of any cell. The system preclassifies to the following WBC classes: Unidentified, Neutrophils, Eosinophils, Lymphocytes, Macrophages (including Monocytes) and Other. Cells preclassified as Basophils, Lymphoma cells, Atypical Lymphocytes, Blasts and Tumor cells are automatically forwarded to the cell class Other. Unidentified is a class for cells and objects which the system has preclassified with a low confidence level.

2. Principles of Operation:
Body fluid samples are prepared by using a cytocentrifuge and a staining unit. Twelve slides can be loaded into each magazine. The magazines are put into the system and the slides are then processed individually. The analysis process consists of an overview image processing and a cell-location step. The body fluid overview image displays the entire sample area. The overview image can be used to find cells of interest and for getting an overall impression of the sample. The overview image can either have one 10x zoom level or both 10x and 50x zoom levels. The cell-location step uses the optical unit and a camera taking images of the identified cells and transferring them to a computer system. The computer system preclassifies the images and stores the images of the located cells and the results in a database, and displays the images in an organized manner.

3. Modes of Operation:
The CellaVision® DM 1200 is an automated cell-locating device. Stained slides are loaded into a magazine with the capacity for 12 slides. The device fits one magazine.

4. Specimen Identification:
The device requires barcode labeled slides for processing.

5. Specimen Sampling and Handling:

{1}

Glass slide with ground or clipped corners containing cytocentrifuged and stained body fluid is analyzed. The cytocetrifugation and staining occurs separately from the device.

6. Calibration: Not applicable
7. Quality Control: The cell location test is used to verify the slide preparation process and the system hardware. In addition, the device performs self-tests during startup of the software, and at certain points during the operation of the system. When the software starts, the system is checked before the operator can start analyses.
8. Software: FDA has reviewed applicant's Hazard Analysis and Software Development processes for this line of product types: Yes ☐ or No ☑

F. Regulatory Information:

1. Regulation section: 21 CFR 864.5260, Automated cell-locating device
2. Classification: Class II
3. Product code: JOY
4. Panel: Hematology (81)

G. Intended Use:

1. Indication(s) for Use: DM1200 is an automated system intended for in-vitro diagnostic use. The body fluid application is intended for differential count of white blood cells. The system automatically locates and presents images of cells on cytocentrifuged body fluid preparations. The operator identifies and verifies the suggested classification of each cell according to type. DM1200 is intended to be used by skilled operators, trained in the use of the device and in recognition of blood cells.
2. Special Conditions for Use Statement(s): Not applicable

H. Substantial Equivalence Information:

1. Predicate Device Name(s) and 510(k) numbers: CellaVision® DM96 with the body fluid application, k080595
2. Comparison with Predicate Device:

| Similarities | | |
| --- | --- | --- |
| Item | DM1200 | DM96 (Predicate) |
| Intended use | Automated cell-locating device for cell-location and identification of white blood cells in cytocentrifuged body fluid preparations, for in-vitro diagnostic use. Verification of results by skilled human operator. | Same |

{2}

| Similarities | | |
| --- | --- | --- |
| Item | DM1200 | DM96 (Predicate) |
| Sample type/Sample preparation | Body fluid samples are prepared using a cytocentrifuge and stained with a Romanowsky stain. | Same |
| Analysis technique | Nucleated cells are located/counted by moving according to the battlement track pattern. Cell images are analyzed using standard mathematical methods, including deterministic artificial neural networks (ANN’s) trained to distinguish between classes of white blood cells. The cell images are preclassified by the system and the operator verifies the suggested classification by either accepting or reclassifying. | Same |
| Overview image | The device presents an overview image. The image gives the operator possibilities to get an overview on parts of or the whole slide in different magnifications. | Same |
| Differences | | |
| --- | --- | --- |
| Item | DM1200 | DM96 (Predicate) |
| Slide handling/ Loading capacity | Slides are loaded into a magazine with the capacity of 12 slides. Device fits one magazine. | Slides are loaded into magazines with the capacity of 12 slides each. Device fits eight magazines. |
| Throughput estimate | 3 to 15 slides per hour for differential | 7 to 25 slides per hour for differential |

I. Special Control/Guidance Document Referenced (if applicable):
EP05-A2, Evaluation of Precision Performance of Quantitative Measurement Methods. Approved Guideline – Second Edition. CLSI
EP09-A2, Method Comparison and Bias Estimation Using Patient Samples; Approved Guideline, Second Edition. CLSI

J. Performance Characteristics:
1. Analytical Performance:
a. Accuracy:
A method comparison of CellaVision® DM 1200 with the body fluid application (test method) and Cellavision® DM96 with the body fluid application (predicate method) was performed. A 200-cell differential leukocyte count was obtained on duplicate slides. The study occurred at two separate clinical sites. The acceptance criteria are as follows: Slope between 0.9 – 1.1 within 95% confidence interval and an intercept between -0.1 - +0.1 within the 95% confidence interval.

{3}

Total samples included in the study.

| Type | Number of samples |
| --- | --- |
| CSF | 62 |
| Serous fluid | 151 |
| Synovial fluid | 47 |
| Total | 260 |

Accuracy results. All samples included (pooled data).

| Cell class | Accuracy | 95% CI Slope | 95% CI Intercept | Number of samples |
| --- | --- | --- | --- | --- |
| Neutrophils | y = 0.9969x + 0.0050 R²= 0.9932 | 0.9868–1.0070 | 0.0004–0.0096 | 260 |
| Lymphocytes | y = 0.9815x + 0.0016 R²= 0.9829 | 0.9656–0.9973 | -0.0049–0.0081 | 260 |
| Eosinophils | y = 1.1048x - 0.0002 R²= 0.9629 | 1.0782–1.1314 | -0.0007–0.0003 | 260 |
| Macrophages | y = 1.0067x - 0.0050 R²= 0.9823 | 0.9901–1.0232 | -0.0125–0.0024 | 260 |
| Other cells | y = 0.9534 + 0.0032 R²= 0.9273 | 0.9207–0.9861 | -0.0002–0.0065 | 260 |

# b. Precision

# Repeatability:

Two samples were prepared in duplicate and labeled sample A and sample B. The evaluation consisted of a 200-cell differential count performed on cytocentrifuged body fluid preparations. The differential count was repeated twenty times $(n = 20)$ using two CellaVision® DM 1200 instruments. Results of the study including acceptance criteria are displayed in the following tables:

Precision results sample A

| Cell Class | Mean % | Sr % within-run | CV % within-run | ST % within-device | CV % within-device | Acceptance Criteria |
| --- | --- | --- | --- | --- | --- | --- |
| Neutrophils | 3.8 | 1.6 | N/A | 1.5 | N/A | ≤ 3% SD |
| Lymphocytes | 29.0 | 2.9 | 10.2 | 3.4 | 11.8 | ≤ 20% CV |
| Macrophages | 63.0 | 2.6 | 4.2 | 3.1 | 3.9 | ≤ 8% CV |
| Other Cells | 1.0 | 0.9 | N/A | 0.8 | N/A | ≤ 3% SD |
| Neutrophils | 6.9 | 1.5 | N/A | 1.4 | N/A | ≤ 3% SD |
| Lymphocytes | 30.1 | 4.2 | 13.9 | 3.9 | 12.9 | ≤ 20% CV |
| Macrophages | 59.0 | 3.5 | 5.9 | 3.5 | 6.0 | ≤ 20% CV |
| Other Cells | 1.0 | 1.0 | N/A | 0.9 | N/A | ≤ 3% SD |

{4}

Precision results sample B

| Cell Class | Mean % | Sr % within-run | CV % within-run | ST % within-device | CV % within-device | Acceptance Criteria |
| --- | --- | --- | --- | --- | --- | --- |
| Neutrophils | 4.2 | 1.2 | N/A | 1.1 | N/A | ≤ 3% SD |
| Lymphocytes | 71.2 | 2.6 | 3.7 | 2.2 | 3.1 | ≤ 8% CV |
| Macrophages | 18.6 | 1.7 | 9.3 | 1.7 | 9.0 | ≤ 20% CV |
| Other Cells | 2.2 | 0.8 | N/A | 0.7 | N/A | ≤ 3% SD |
| Neutrophils | 3.3 | 1.6 | N/A | 1.2 | N/A | ≤ 3% SD |
| Lymphocytes | 72.4 | 1.9 | 2.6 | 1.6 | 2.2 | ≤ 8% CV |
| Macrophages | 20.2 | 1.7 | 8.7 | 1.4 | 7.0 | ≤ 20% CV |
| Other Cells | 1.8 | 0.9 | N/A | 0.7 | N/A | ≤ 3% SD |

## Reproducibility

A reproducibility study was conducted at three sites using three CellaVision® DM 1200 instruments with each site processing the same twelve samples in duplicate, for 20 operating days (n = 120 per sample). All results of the study demonstrate that the precision of the CellaVision® DM 1200 between different devices, sites, operators, and days is within the acceptance criteria listed in the following table:

Precision acceptance criteria (reproducibility)

| Differential value | Acceptance criteria |
| --- | --- |
| 0-10 % | ≤ 3% SD |
| 11-60 % | ≤ 20% CV |
| 61-100 % | ≤ 8% CV |

c. Linearity: Not applicable
d. Carryover: Not applicable
e. Interfering Substances: Not applicable

2. Other Supportive Instrument Performance Data Not Covered Above: none

## K. Proposed Labeling:

The labeling is sufficient and it satisfies the requirements of 21 CFR Part 809.10.

## L. Conclusion:

The submitted information in this premarket notification is complete and supports a substantial equivalence decision.