(300 days)
The HemoScreen is a point-of-care (POC) automated hematology analyzer intended for the enumeration and classification of the following parameters in capillary and venous whole blood (K2EDTA anticoagulated): WBC, RBC, HGB, HCT, MCV, MCH, MCHC, RDW, PLT, MPV, NEUT%, NEUT#, LYMP%, LYMP#, MONO%, MONO#, E0%, E0%, BASO %, and BASO#. The HemoScreen is for in vitro diagnostic use in clinical laboratories and/or POC settings for adults and children at least 2 years of age.
HemoScreen is a point of care (POC), automated hematology analyzer that provides 20 common CBC parameters, including a 5-part leukocyte (WBC) differential, in capillary and venous whole blood samples. The HemoScreen analyzer (reader) is a tabletop device that is designed to use with a disposable reagent cartridge. In addition to the cartridge, the system includes a disposable sampler with two glass capillaries which is used to collect the blood sample and then transfer it to the cartridge. Once the cartridge is inserted into the reader, there are no further procedural steps; blood is expelled from the capillaries (sampler) into the reagent compartments (cartridge). The reader then mixes the blood sample with the reagents by alternately pressing compressible portions of the cartridge, eventually causing the suspension of cells to flow into the microfluidic chamber. Cells flowing in the microfluidic chamber focus into a single-cell plane due to a patented physical phenomenon known as viscoelastic focusing. The reader then captures images of the focused cells and analyzes them in real time using machine vision algorithms. When analysis is complete, the results are displayed to the user on the reader's touch screen and may be printed to an adjacent printer or exported to a USB flash drive. The cartridge is ejected by the analyzer after analysis, and can then be safely disposed of, as the reagents and blood sample remain within the cartridge. The basic staining and microscopic image analysis performed by HemoScreen closely resembles the traditional blood smear and the hemocytometer counting chamber. Leukocytes are classified based on their staining properties and morphology, whereas absolute counts are obtained by counting the cells contained in a chamber of predetermined volume. Test results are obtained within six (6) minutes and the results are saved.
Here's a breakdown of the acceptance criteria and study information for the HemoScreen Hematology Analyzer based on the provided text:
Acceptance Criteria and Device Performance
The acceptance criteria are generally implied by the statement "the redefined acceptance criteria were met for all the 20 measurands and in all tested ranges" for precision, and "the 95% confidence intervals for mean bias / mean relative bias were within the acceptance limits" and "the correlation point estimates met the acceptance criteria" for method comparison. Specific numerical acceptance limits are not explicitly stated in the provided text for all parameters (e.g., a specific Pearson correlation threshold, or a specific mean bias range), but the tables present the reported device performance.
Table of Acceptance Criteria (Implied) and Reported Device Performance (Focus on Method Comparison data, as this provides a clearer standard for comparison):
| Parameter (Units) | Implied Acceptance Criteria (e.g., acceptable Pearson Correlation, acceptable Mean Bias) | Reported Pearson Correlation (r) vs. Sysmex | Reported Mean Bias vs. Sysmex | Reported Pearson Correlation (r) vs. Blood Smear | Reported Mean Bias vs. Blood Smear |
|---|---|---|---|---|---|
| WBC Parameters | |||||
| WBC (x 10³/μL) | Within acceptance limits | 0.993 | 0.09 ( -0.14%) | N/A | N/A |
| NEUT (x 10³/μL) | Within acceptance limits | 0.989 | 0.22 (3.83%) | N/A | N/A |
| LYM (x 10³/μL) | Within acceptance limits | 0.997 | 0.12 (5.46%) | N/A | N/A |
| MON (x 10³/μL) | Within acceptance limits | 0.718 | -0.16 (-27.08%) | N/A | N/A |
| EOS (x 10³/μL) | Within acceptance limits | 0.981 | -0.00 (-4.33%) | N/A | N/A |
| BAS (x 10³/μL) | Within acceptance limits (BASO Study) | 0.484 | 0.04 (11.87%) | N/A | N/A |
| NEUT (%) | Within acceptance limits | 0.984 | 1.90 (3.71%) | 0.960 | -0.30 (-0.23%) |
| LYM (%) | Within acceptance limits | 0.991 | 0.96 (5.11%) | 0.961 | 1.01 (8.45%) |
| MON (%) | Within acceptance limits | 0.780 | -1.87 (-26.90%) | 0.594 | -0.17 (-2.41%) |
| EOS (%) | Within acceptance limits | 0.981 | -0.06 (-2.97%) | 0.943 | 0.13 (16.01%) |
| BAS (%) | Within acceptance limits (BASO Study) | N/A | N/A | 0.725 | -0.01 (43.99%) |
| RBC Parameters | |||||
| RBC (x 10⁶/μL) | Within acceptance limits | 0.989 | 0.02 (0.52%) | N/A | N/A |
| HGB (g/dL) | Within acceptance limits | 0.985 | 0.13 (0.81%) | N/A | N/A |
| HCT (%) | Within acceptance limits | 0.982 | 1.15 (3.05%) | N/A | N/A |
| MCV (fL) | Within acceptance limits | 0.956 | 2.27 (2.53%) | N/A | N/A |
| MCH (pg) | Within acceptance limits | 0.957 | 0.06 (0.20%) | N/A | N/A |
| MCHC (g/dL) | Within acceptance limits | 0.675 | -0.78 (-2.34%) | N/A | N/A |
| RDW (%) | Within acceptance limits | 0.946 | -0.03 (-0.12%) | N/A | N/A |
| PLT Parameters | |||||
| PLT (x 10³/μL) | Within acceptance limits | 0.967 | 14.57 (5.12%) | N/A | N/A |
| MPV (fL) | Within acceptance limits | 0.831 | 0.21 (1.93%) | N/A | N/A |
Note on Basophils (BAS# and BAS%): The initial method comparison notes that basophil data were "not meaningful" due to the distribution of samples. A separate "BASO Study" was conducted for basophils, which yielded specific correlation and bias values as shown.
Study Information:
1. Sample sizes used for the test set and the data provenance:
- Method Comparison & Clinical Sensitivity/Specificity:
- Sample Size: 495 normal and pathological residual whole blood specimens.
- Data Provenance: Collected across three clinical sites (retrospective, as they were "residual" samples).
- Basophil-specific study ("BASO Study"):
- Sample Size: 95 whole blood samples that included high levels of basophils.
- Data Provenance: Not explicitly stated if it's from the same three clinical sites or if it's retrospective/prospective, but likely similar to the main method comparison.
- Flagging Study:
- Sample Size: 402 whole blood specimens.
- Data Provenance: Analyzed across three clinical sites (retrospective, as samples were "analyzed").
- Reference Intervals:
- Sample Size: Minimum of 120 male subjects and 120 female subjects (total 243 subjects reported: 123 female, 120 male).
- Data Provenance: Single US site, freshly collected K2EDTA venous blood from healthy (self-reported) adult (19-69 years old) male and female volunteers on a single occasion (prospective).
- Vein to Capillary Equivalency:
- Sample Size: 75 normal and pathological paired capillary and venous whole blood specimens.
- Data Provenance: Drawn from volunteer subjects across three clinical sites (prospective, as samples were "drawn").
2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:
- Method Comparison & Clinical Sensitivity/Specificity, Flagging, Vein to Capillary Equivalency Studies:
- For blood smear analysis (part of the ground truth for differentials and flagging), "Experienced operators trained to use the automated blood smear performed the differentials, including morphology evaluation, and the results were verified by these operators." The exact number or specific qualifications (e.g., "radiologist with 10 years of experience") are not provided, other than "experienced operators."
- For the comparative instrument (Sysmex XN Series), "laboratory personnel" performed the analyses. Specific qualifications are not detailed.
- BASO Study (ground truth for basophils):
- Light microscopy (standard reference method) was used for %basophils, performed by "experienced operators" (implied, similar to other blood smear analyses).
- Sysmex method was used for absolute counts, performed by "laboratory personnel."
3. Adjudication method (e.g., 2+1, 3+1, none) for the test set:
- The document mentions that for blood smear analysis, "the results were verified by these operators," suggesting internal verification rather than a formal multi-expert adjudication method like 2+1 or 3+1. It does not explicitly state a formal adjudication process for discordant results between the device and the ground truth.
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, a MRMC comparative effectiveness study involving human readers improving with/without AI assistance was not done. The HemoScreen is an automated hematology analyzer, not an AI-assisted diagnostic tool for human readers. The clinical studies compare its performance against established laboratory methods (Sysmex) and manual blood smear analysis, not evaluating human reader performance.
5. If a standalone (i.e. algorithm only without human-in-the loop performance) was done:
- Yes, the studies evaluate the standalone performance of the HemoScreen Hematology Analyzer. The device is an automated system that performs enumeration and classification using its internal machine vision algorithms. The "method comparison" sections directly assess the algorithm's performance against reference methods.
6. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):
- For enumeration parameters (WBC, RBC, HGB, PLT etc.): The ground truth was primarily established by a legally marketed predicate device, the Sysmex XN Series, and in some cases, by other reference methods like centrifugation for LoB/LoD/LoQ studies.
- For differential parameters (NEUT%, LYM%, MONO%, EOS%, BASO%): The ground truth was established by the predicate device (Sysmex XN Series) and manual blood film differential counts (microscopy) performed by "experienced operators trained to use the automated blood smear," following CLSI H20-A2 guidelines (2 slides / 200 cells counted per slide for a total of 400 cells). This falls under a form of expert consensus/manual review.
- For Flagging: Ground truth was based on differential results from the predicate and manual blood smears.
- For LoB/LoD/LoQ: Processed blood samples measured by a "comparative method" (likely a reference method capable of precise low-level measurement).
7. The sample size for the training set:
- The document does not explicitly state the sample size for the training set. The provided information pertains to the validation studies (nonclinical and clinical data proving performance). Automated analyzers like HemoScreen employ machine vision algorithms, which would have been trained on a separate dataset, but the details of this training dataset are not included in the 510(k) summary.
8. How the ground truth for the training set was established:
- As the training set size is not provided, the method for establishing its ground truth is also not described in this document. For such devices, ground truth for training would typically involve large datasets of images with expert-verified cell classifications and counts.
§ 864.5220 Automated differential cell counter.
(a)
Identification. An automated differential cell counter is a device used to identify one or more of the formed elements of the blood. The device may also have the capability to flag, count, or classify immature or abnormal hematopoietic cells of the blood, bone marrow, or other body fluids. These devices may combine an electronic particle counting method, optical method, or a flow cytometric method utilizing monoclonal CD (cluster designation) markers. The device includes accessory CD markers.(b)
Classification. Class II (special controls). The special control for this device is the FDA document entitled “Class II Special Controls Guidance Document: Premarket Notifications for Automated Differential Cell Counters for Immature or Abnormal Blood Cells; Final Guidance for Industry and FDA.”