Report

Swelab™ Lumi automated hematology analyzer

Performance validation
Contents

1. Introduction 3 8. Analytical specificity 11

8.1. Samples 11
2. Swelab Lumi description 3 8.2. Test procedure 11
8.3. Test results 11
3. Evaluation 3
3.1 Validated performance 3 9. WBC classification –
3.2. Applicable standards and guidelines 4 clinical sensitivity and specificity 11
9.1. Samples 11

4. Preparations 4 9.2. Test procedure 11

9.3. Test results 12

5. Trueness 4
5.1. Test procedure 4 10. Carryover 12

5.2 Test results 5 10.1. Samples 12

10.2. Test procedure 12

6. Precision 6 10.3. Test results 12

6.1. Repeatability 6
6.1.1. Samples 6 11. Reference intervals 13

6.1.2. Test procedure 6 11.1. Samples 13

6.1.3. Test results 6 11.2. Test procedure 13

6.2 Reproducibility 6 11.3. Test results 13

6.2.1. Samples 6
6.2.2. Test procedure 6 12. Analytical measuring ranges
(linearity ranges) 14
6.2.3. Test results 6
12.1. Samples 14
12.2. Test procedure 14
7. Analytical sensitivity 9
12.3. Test results 15
7.1. Limit of blank (background) 9
7.1.1. Samples 9
13. Mode-to-mode comparability 16
7.1.2. Test procedure 9
7.1.3. Test results 10
7.2. Limit of detection 10
7.2.1. Samples 10
7.2.2. Test procedure 10
7.2.3. Test results 10
7.3. Limit of quantitation 10
7.3.1. Samples 10
7.3.2. Test procedure 10
7.3.3. Test results 10

2
1. Introduction
Laboratory diagnostics is one of the cornerstones of healthcare, and test results form the basis for patient
diagnosis. A complete blood count (CBC) is typically the first test requested by a physician to evaluate a
patient’s general health status. An automated hematology system that is used for such testing constitutes
a cost-efficient tool for initial disease investigation and follow-up. It is therefore of utmost importance that
the reported results are analytically valid.

Automated hematology analyzers are both analytically and technically complex. The analysis is performed
in parallel for a range of parameters that concern blood cell number and size as well as hemoglobin
concentration. According to regulatory requirements, diagnostic systems are validated by the manufacturer
before released to the market. International organizations such as the Clinical and Laboratory Standards
Institute (CLSI) provide guidelines for the verification items. Several international directives and regulations
dictate that each laboratory verifies the performance of diagnostic systems intended to be used in routine
clinical care.

This document compiles the clinical performance validation of Boule Swelab Lumi hematology system. The
studies were performed on behalf of Boule at a clinical site by experienced laboratory staff and operators.
The aim of this report is to support diagnostic laboratories in verifying the analytical performance of
Swelab Lumi before taking the system into routine use.

2. Swelab Lumi description

Swelab Lumi is a quantitative, multi-parameter automated hematology analyzer intended for blood
cell counting and sizing for in vitro diagnostic use in screening of human capillary or venous whole blood
samples collected in K2EDTA blood collection tubes to collect data reflecting the patient’s hematological
status. The analyzer must be used together with Swelab Lumi-D Diluent, L1 Lyse and L2 Lyse reagents.

The device is intended for laboratory professional use.

The Swelab Lumi is used for enumeration of white blood cells (WBC); the absolute number and percentage
of neutrophils (NEU#, NEU%), lymphocytes (LYM#, LYM%), monocytes (MON#, MON%), eosinophils (EOS#,
EOS%), basophils (BAS#, BAS%); red blood cells (RBC); hemoglobin concentration (HGB); hematocrit (HCT);
mean corpuscular volume (MCV); mean corpuscular hemoglobin (MCH); mean corpuscular hemoglobin
concentration (MCHC); red blood cell distribution width coefficient of variation (RDW-CV); platelet count
(PLT); mean platelet volume (MPV).

3. Evaluation
3.1 Validated performance
This document comprises validation of trueness, precision (repeatability and reproducibility), analytical
sensitivity (LoB, LoD, LoQ), analytical specificity (interferences), carryover, reference intervals, analytical
measuring ranges (linearity ranges), and mode-to-mode comparability.

The Swelab Lumi hematology analyzer and its associated reagents, calibrators, and control materials were
used as the test device. Corresponding components of a reference system were used where applicable.
Manual microscopy was used as reference method where applicable. Fresh whole blood samples used in
the validation testing were collected for routine analysis in K2EDTA blood collection tubes.

3
3.2. Applicable standards and guidelines
H26-A2 Validation, verification, and quality assurance of automated hematology analyzers; 2010
Approved standard-second edition
EP17-A2 Evaluation of Detection Capability for Clinical Laboratory Measurement Procedures; 2012
approved guideline - second edition
EP05-A3 Evaluation of precision of quantitative Measurement Procedures; approved guideline 2014
- Third edition
EP06-Ed2 Evaluation of linearity of quantitative Measurement Procedures 2020
EP09-A3 Measurement Procedure Comparison and Bias estimation using patient samples; 2013
approved guideline- Third edition
EP28-A3c Defining, Establishing, and verifying reference intervals in the clinical laboratory; 2010
approved guideline- Third edition
WST 405-2012 Ministry of Health of the People’s Republic of China. Reference intervals for blood cell 2012
analysis. Beijing. Standards Press of China.
WST 779–2021 Ministry of Health of the People’s Republic of China. The reference interval for blood 2021
cell analysis in children. Beijing. Standard Press of China.
EP07,3rd ed. Interference testing in clinical chemistry 2018
H20-A2 Reference Leukocyte (WBC) differential count (proportional) and evaluation of 2007
instrumental methods; Approved Standards- second edition
ICSH ICSH guidelines for the evaluation of blood cell analysers including those used 2014
for differential leucocyte and reticulocyte counting. International Council for
Standardization in Hematology.
literature Int. Jnl. Lab. Hem. 2014, 36, 613-627

4. Preparations
Before starting a test, the following preparations were performed:
1. Instrument start-up and routine cleaning procedures were performed as defined in the User manual.
2. Calibration of the instrument was verified.
3. A background count was performed to check that the results were well below the acceptance values
stated in the User manual.
4. Con 5-Diff Low, Normal, and High controls were used to check that the results were within the assay
sheet values.

5. Trueness
Fresh anticoagulated venous whole blood samples (n = 100) were analyzed in the venous blood mode on
the test system and the reference system, respectively. Samples included in the study were selected to
support the main test parameters WBC, RBC, HGB, MCV, PLT.

The following samples were excluded:

• 
Samples including microcytes
• 
Samples including nucleated red blood cells
• 
Samples including immature cells

5.1. Test procedure

The XN-1000 5-part hematology analyzer and associated reagents, calibrator, and control material
(Sysmex Corp.) were used as reference system. The test and reference systems were operated according
to instructions for performing the cell count in their respective user manual. The strength of the relationship
between the cell count in the test and the reference systems was measured using Pearson correlation
coefficient (r). Passing-Bablok regression analysis and Bland-Altman difference plots were performed on
matched samples for estimation of agreement and possible systematic bias between the test and the
reference systems.

4
5.2 Test results
Correlation results, with specification limits for the correlation coefficient (r) and bias between test and
reference systems, are given in Table 5-1. Correlation plots are shown in Figure 5-1. As shown from the study
results, Swelab Lumi performance fulfills the specifications for trueness.

Table 5-1. Correlation of Swelab Lumi versus reference system

Result/specification WBC RBC HGB MCV PLT

r 0.999 0.998 0.998 0.981 0.994
r, specification ≥ 0.99 ≥ 0.99 ≥ 0.98 ≥ 0.98 ≥ 0.95
Bias (%) 0.25 0.88 0.35 1.83 1.79
Bias, specification ± 5% ± 3% ± 2.5% ± 3% ± 7%

(A) (B)
35 7
y = 1.0023x − 0.0313 y = 0.9811x + 0.1104
30 R2 = 0.9994 6 R2 = 0.995
Swelab Lumi: WBC (109/L)

Swelab Lumi: RBC (1012/L)

25 5

20 4

15 3

10 2

5 1

0 0
0 5 10 15 20 25 30 35 0 1 2 3 4 5 6 7
Sysmex: WBC (109/L) Sysmex: RBC (1012/L)
(C) (D)
200 120
y = 1.0116x − 0.9427 y = 0.9673x + 1.3254
R2 = 0.9975 100 R2 = 0.9628
Swelab Lumi: HGB (g/dL)

150
Swelab Lumi: MCV (fL)

80

100 60

40
50
20

0 0
0 50 100 150 200 0 20 40 60 80 100 120
Sysmex: HGB (g/dL) Sysmex: MCV (fL)
(E)
1000
y = 0.9863x + 5.5968
R2 = 0.9886
800
Swelab Lumi: PLT (109/L)

600

400

200

0
0 200 400 600 800 1000
Sysmex: PLT (109/L)

Fig 5-1. Correlation plots for (A) WBC, (B) RBC, (C) HGB, (D) MCV, and (E) PLT. In the regression plots, the dotted line
corresponds to best fit.

5
6. Precision
Precision is a measure of the agreement between repeated measurements of a sample and is expressed
as coefficient of variation (CV%).

6.1. Repeatability
Repeatability (short term precision) is a measure of precision under the same conditions, meaning testing
is performed by the same operator, using the same analyzer and the same set of reagents.

6.1.1. Samples
Fresh normal (three concentrations) and abnormal (three concentrations) human whole blood samples
were analyzed in 10 consecutive replicates.

6.1.2. Test procedure

Ten replicates were analyzed for each of the six samples on the three analyzers.

6.1.3. Test results

The mean, SD, and CV% were calculated for each sample. As shown from Table 6-1, repeatability results
(short term precision) for the Swelab Lumi system are all within acceptance criteria for the normal range of
the tested parameters.

6.2 Reproducibility
Reproducibility (long term precision) is a measure of precision when the study is reproduced in its entirety.

6.2.1. Samples
One set of controls:
• 
Con 5-Diff Normal
• 
Con 5-Diff Low
• 
Con 5-Diff High

6.2.2. Test procedure

The same set of controls (same lot number) was tested in duplicate on three analyzers that were set up
with different sets of reagents, and data was collected for WBC, RBC, HGB, HCT, and PLT from two runs
per day for 20 days.

6.2.3. Test results

The mean, SD, and CV values for each control level were obtained from pooled data from the three
analyzers. Imprecision of the analyzers, expressed as SD and CV values, was calculated for:
• 
Within run (Sr) = between duplicate tests
• 
Between run (Srr) = between duplicate tests in the morning and in the afternoon
• 
Between day (Sdd) = between the 20 days
• 
Within device (ST) = all above (total imprecision)

As shown from Table 6-2, the reproducibility results (long term precision) for the Swelab Lumi system were
within the acceptance criteria for all parameters, for all three levels of controls.

6
Table 6-1. Repeatability of Swelab Lumi

Device 1 WBC NEU% LYM% MON% EOS% BAS% RBC HGB MCV PLT MPV
Sample 1
Mean 4.80 55.5 34.5 7.4 2.1 0.6 4.99 151 90.3 158 9.9
CV%/SD 1.9% -1.4 0.6 1.1 -0.3 -0.4 1.1% 0.7% 0.2% 2.9% 1.7%
Sample 2
Mean 7.29 58.2 33.1 6.1 2.1 0.6 4.50 133 88.7 266 9.3
CV%/SD 1.2% -1.3 -0.9 -0.8 -0.1 -0.1 0.6% 0.5% 0.2% 1.1% 1.8%
Sample 3
Mean 6.06 56.7 34.3 6.0 2.2 0.8 4.80 148 90.7 242 10.1
CV%/SD 1.8% -1.6 1.1 0.8 -0.3 -0.2 0.7% 0.5% 0.1% 2.2% 1.3%
Sample 4
Mean 1.20 25.30 58.29 10.14 4.98 1.29 3.72 114.80 89.57 75.80 9.99
CV%/SD 3.2% 2.6 -2.4 3.1 1.6 -0.6 0.5% 0.5% 0.1% 3.8% 2.2%
Sample 5
Mean 3.51 63.41 26.05 8.60 1.48 0.46 2.62 84.30 93.44 33.60 9.34
CV%/SD 1.9% 2.1 1.2 2.1 0.5 -0.2 0.4% 0.8% 0.1% 8.3% 2.5%
Sample 6
Mean 1.05 10.15 84.37 4.58 0.17 0.73 2.23 71.60 96.91 200.90 8.23
CV%/SD 3.2% -2.6 -2.3 1.3 -0.2 -0.7 0.5% 0.9% 0.1% 1.6% 2.2%
Acceptance ≤ 2.5% ± 4.0 SD ± 3.0 SD ± 2.0 SD ± 1.5 SD ± 0.8 SD ≤ 1.5% ≤ 1.5% ≤ 1.0% ≤ 6.0% (100- ≤ 4.0
%CV/SD (4.0-15.0 (50.0%- (25.0%- (5.0%-10.0%) (2.0%-5.0%) (0.5%-1.5%) (3.5-6.0 (110-180 g/L) (70-120 fL) 149 × 109/L)
× 109/L) 60.0%) 35.0%) × 1012/L) ≤ 4.0% (150-
500 × 109/L)

Device 2 WBC NEU% LYM% MON% EOS% BAS% RBC HGB MCV PLT MPV
Sample 1
Mean 4.78 54.9 35.1 7.2 2.1 0.6 4.95 151 90.3 158 10.0
CV%/SD 1.4% -1.8 -1.4 -1.1 -0.3 -0.4 1.2% 0.6% 0.2% 2.3% 2.8%
Sample 2
Mean 7.31 58.0 33.1 6.1 2.1 0.6 4.50 132 88.8 266 9.3
CV%/SD 0.9% -1.1 0.6 0.5 0.1 -0.1 0.8% 0.8% 0.1% 1.2% 1.4%
Sample 3
Mean 7.11 56.3 34.5 5.8 2.5 0.9 4.46 128 85.1 278 11.3
CV%/SD 1.2% -0.3 0.3 -0.2 0.4 -0.1 0.6% 0.3% 0.2% 2.0% 1.3%
Sample 4
Mean 1.09 10.17 84.22 4.64 0.24 0.73 2.27 73.10 97.14 200.90 8.23
CV%/SD 4.3% -2.3 -2.0 1.1 -0.4 -0.8 1.7% 0.7% 0.1% 1.6% 2.2%
Sample 5
Mean 3.55 63.61 26.00 8.38 1.46 0.55 2.68 85.80 93.62 33.60 9.34
CV%/SD 2.1% 2.1 1.2 1.9 0.5 0.3 1.0% 1.1% 0.2% 8.3% 2.5%
Sample 6
Mean 1.25 25.20 58.45 10.05 5.01 1.29 3.77 116.30 89.80 75.80 9.99
CV%/SD 3.0% 2.8 -2.3 3.3 1.9 -0.5 0.9% 0.8% 0.2% 3.8% 2.2%
Acceptance ≤ 2.5% ± 4.0 SD ± 3.0 SD ± 2.0 SD ± 1.5 SD ± 0.8 SD ≤ 1.5% ≤ 1.5% ≤ 1.0% ≤ 6.0% (100- ≤ 4.0
%CV/SD (4.0-15.0 (50.0%- (25.0%- (5.0%-10.0%) (2.0%-5.0%) (0.5%-1.5%) (3.5-6.0 (110-180 g/L) (70-120 fL) 149 × 109/L)
× 109/L) 60.0%) 35.0%) × 1012/L) ≤ 4.0% (150-
500 × 109/L)

Device 3 WBC NEU% LYM% MON% EOS% BAS% RBC HGB MCV PLT MPV
Sample 1
Mean 6.89 55.7 34.6 6.4 2.5 0.9 4.76 149 93.5 273 8.9
CV%/SD 1.0% 0.5 0.3 0.2 0.3 -0.1 0.4% 0.5% 0.2% 1.1% 1.8%
Sample 2
Mean 5.15 55.0 34.8 5.9 3.4 1.0 4.85 154 96.1 215 9.3
CV%/SD 1.7% -0.9 -1.3 0.9 -0.4 -0.3 0.6% 0.3% 0.1% 1.9% 1.5%
Sample 3
Mean 5.14 55.2 34.6 6.6 2.9 0.6 4.19 124 88.6 215 9.3
CV%/SD 1.9% -0.5 0.3 0.3 0.2 0.1 0.6% 0.7% 0.1% 1.9% 1.5%
Sample 4
Mean 1.15 10.11 84.16 4.62 0.35 0.76 2.31 74.70 97.46 200.90 8.30
CV%/SD 4.7% -2.4 -2.0 1.1 -0.5 -0.8 2.6% 1.0% 0.1% 1.6% 2.3%
Sample 5
Mean 3.61 63.90 25.59 8.38 1.54 0.59 2.72 87.40 93.84 33.60 9.39
CV%/SD 1.7% 2.2 2.8 2.0 -0.4 -0.3 1.2% 1.3% 0.2% 8.3% 2.3%
Sample 6
Mean 1.32 25.17 58.49 9.98 5.02 1.34 3.83 117.80 89.99 75.80 10.06
CV%/SD 2.4% 3.1 -2.0 3.2 1.8 -0.5 1.4% 0.8% 0.2% 3.8% 2.2%
Acceptance ≤ 2.5% ± 4.0 SD ± 3.0 SD ± 2.0 SD ± 1.5 SD ± 0.8 SD ≤ 1.5% ≤ 1.5% ≤ 1.0% ≤ 6.0% (100- ≤ 4.0
%CV/SD (4.0-15.0 (50.0%- (25.0%- (5.0%-10.0%) (2.0%-5.0%) (0.5%-1.5%) (3.5-6.0 (110-180 g/L) (70-120 fL) 149 × 109/L)
× 109/L) 60.0%) 35.0%) × 1012/L) ≤ 4.0% (150-
500 × 109/L)

7
Table 6-2. Reproducibility of Swelab Lumi

WBC
Device Analyzer 1 Analyzer 2 Analyzer 3
Control Low Normal High Low Normal High Low Normal High
Mean 3.67 8.24 18.75 3.89 8.78 19.24 3.80 8.70 19.17
Sr 0.0719 0.1050 0.3975 0.1160 0.3129 0.4057 0.0995 0.3150 0.4073
CV 1.96% 1.27% 2.12% 2.98% 3.56% 2.11% 2.61% 3.62% 2.12%
Acceptance ≤ 3.75%
Sdd 0.1168 0.2494 0.5143 0.0764 0.2542 0.5057 0.0931 0.2541 0.5103
CV 3.18% 3.03% 2.74% 1.96% 2.90% 2.63% 2.45% 2.92% 2.66%
Srr 0.0489 0.0694 0.1636 0.1545 0.0452 0.2630 0.1319 0.0322 0.2663
CV 1.33% 0.84% 0.87% 3.97% 0.51% 1.37% 3.47% 0.37% 1.39%
Acceptance ≤ 5.00%
ST 0.1456 0.2793 0.6703 0.2078 0.4057 0.6997 0.1896 0.4059 0.7051
CV 3.97% 3.39% 3.58% 5.33% 4.62% 3.64% 4.98% 4.67% 3.68%
Acceptance ≤ 7.5%
RBC
Device Analyzer 1 Analyzer 2 Analyzer 3
Control Low Normal High Low Normal High Low Normal High
Mean 2.51 4.79 5.45 2.56 4.84 5.50 2.56 4.84 5.50
Sr 0.0284 0.0297 0.0566 0.0346 0.0426 0.0615 0.0366 0.0483 0.0576
CV 1.13% 0.62% 1.04% 1.36% 0.88% 1.12% 1.43% 1.00% 1.05%
Acceptance ≤ 1.50%
Sdd 0.0273 0.0547 0.0757 0.0297 0.0539 0.0844 0.0244 0.0649 0.0785
CV 1.09% 1.14% 1.39% 1.16% 1.11% 1.53% 0.95% 1.34% 1.43%
Srr 0.0141 0.0281 0.0349 0.0137 0.0221 0.0296 0.0219 0.0169 0.0410
CV 0.56% 0.59% 0.64% 0.54% 0.46% 0.54% 0.86% 0.35% 0.75%
Acceptance ≤ 2.00%
ST 0.0419 0.0683 0.1007 0.0476 0.0721 0.1086 0.0491 0.0791 0.1056
CV 1.67% 1.43% 1.85% 1.86% 1.49% 1.97% 1.92% 1.64% 1.92%
Acceptance ≤ 3.00%
HGB
Device Analyzer 1 Analyzer 2 Analyzer 3
Control Low Normal High Low Normal High Low Normal High
Mean 63.71 141.05 179.46 65.18 142.49 181.09 65.26 142.49 180.94
Sr 0.7984 0.8660 2.0646 1.0247 0.9811 2.1125 0.9811 1.0308 2.0585
CV 1.25% 0.61% 1.15% 1.57% 0.69% 1.17% 1.50% 0.72% 1.14%
Acceptance ≤ 1.75%
Sdd 0.5770 1.4929 2.5791 0.6238 1.4273 2.5657 0.5456 1.5106 2.6271
CV 0.91% 1.06% 1.44% 0.96% 1.00% 1.42% 0.84% 1.06% 1.45%
Srr 0.3708 0.9552 1.2990 0.1118 0.9682 1.3086 0.2958 1.0897 1.1402
CV 0.58% 0.68% 0.72% 0.17% 0.68% 0.72% 0.45% 0.76% 0.63%
Acceptance ≤ 2.33%
ST 1.0526 1.9726 3.5499 1.1944 1.9842 3.5718 1.1609 2.1288 3.5269
CV 1.65% 1.40% 1.98% 1.83% 1.39% 1.97% 1.78% 1.49% 1.95%
Acceptance ≤ 3.50%

8
Continuation of Table 6-2.

HCT
Device Analyzer 1 Analyzer 2 Analyzer 3
Control Low Normal High Low Normal High Low Normal High
Mean 19.76 43.39 54.62 20.18 44.00 55.24 20.18 43.92 55.20
Sr 0.2179 0.2848 0.5878 0.2706 0.4161 0.6387 0.2853 0.4194 0.6296
CV 1.10% 0.66% 1.08% 1.34% 0.95% 1.16% 1.41% 0.95% 1.14%
Acceptance ≤ 1.50%
Sdd 0.1708 0.4174 0.6693 0.1972 0.4069 0.7675 0.1252 0.4239 0.6947
CV 0.86% 0.96% 1.23% 0.98% 0.92% 1.39% 0.62% 0.97% 1.26%
Srr 0.1378 0.2850 0.4164 0.1445 0.2452 0.3767 0.2145 0.3148 0.4511
CV 0.70% 0.66% 0.76% 0.72% 0.56% 0.68% 1.06% 0.72% 0.82%
Acceptance ≤ 2.00%
ST 0.3093 0.5802 0.9833 0.3647 0.6315 1.0672 0.3782 0.6743 1.0404
CV 1.57% 1.34% 1.80% 1.81% 1.44% 1.93% 1.87% 1.54% 1.88%
Acceptance ≤ 3.00%
PLT
Device Analyzer 1 Analyzer 2 Analyzer 3
Control Low Normal High Low Normal High Low Normal High
Mean 63.90 226.71 423.45 62.46 225.18 424.38 62.36 225.24 424.48
Sr 3.4132 7.6018 26.2826 3.4077 7.6420 12.9296 3.4077 7.6820 12.9875
CV 5.34% 3.35% 6.21% 5.46% 3.39% 3.05% 5.46% 3.41% 3.06%
Acceptance ≤ 6.25%
Sdd 1.3099 15.7743 19.7233 1.3318 15.9550 21.3609 1.2455 15.7256 21.4075
CV 2.05% 6.96% 4.66% 2.13% 7.09% 5.03% 2.00% 6.98% 5.04%
Srr 2.6029 8.0893 9.4353 2.6739 7.9930 10.3193 2.7203 7.9357 10.1409
CV 4.07% 3.57% 2.23% 4.28% 3.55% 2.43% 4.36% 3.52% 2.39%
Acceptance ≤ 8.33%
ST 4.4878 19.2887 34.1878 4.5317 19.4126 27.0176 4.5347 19.2167 27.0147
CV 7.02% 8.51% 8.07% 7.26% 8.62% 6.37% 7.27% 8.53% 6.36%
Acceptance ≤ 12.5%

7. Analytical sensitivity
Limit of blank (LoB), limit of detection (LoD), and limit of quantitation (LoQ) are used to describe the
analytical sensitivity, that is, the lowest concentration of a parameter that can be measured in a reliable
manner. For hematology, only WBC and PLT are of clinical interest in the very low range, as near-zero levels
of RBC and HGB are incompatible with human life.

7.1. Limit of blank (background)

LoB refers to the upper limit on expected blank sample measurements and corresponds with what
is traditionally called background. LoB is determined as a first step and is also used to define the
concentration interval of samples for the LoD test.

7.1.1. Samples
On each of three days of testing, four blank samples (laboratory pure water, commercial saline, diluent A,
and diluent B) were tested five times (60 analyses).

7.1.2. Test procedure

To determine LoB, the blank samples were analyzed with two different reagent lots. On each of three days,
5 replicates of one sample were analyzed with each reagent pair. The LoB for WBC, RBC, HGB, and PLT
was determined for the pooled data by non-parametric analysis with 95% confidence.

9
7.1.3. Test results
LoB test results are shown in Table 7-1.

Table 7-1. LoB of Swelab Lumi

Parameter WBC RBC HGB PLT

LoB 0.07 × 109/L 0.00 × 1012/L 0 g/L 0 × 109/L

7.2. Limit of detection

LoD is the lowest amount of parameters in a sample that can be detected with a stated probability, which
in hematology refers to the numerical value for the lowest concentration of a particular blood cell type
that can be discriminated from background.

7.2.1. Samples

On each of three days of testing, four low-concentrate samples (ranging from 1 to 5 times LoB) were tested
five times (60 analyses).

7.2.2. Test procedure

To determine LoD, the samples were analyzed with two different reagent lots. On each of three days, 5
replicates of one sample were analyzed with each reagent pair. The LoD for WBC and PLT was determined
for the pooled data by non-parametric analysis with 95% confidence.

7.2.3. Test results

LoD results are shown in Table 7-2.

Table 7-2. LoD of Swelab Lumi

Parameter WBC PLT

LoD 0.13 × 10 /L
9
1.41 × 109/L

7.3. Limit of quantitation

LoQ is the lowest amount of a parameter in a sample that can be quantifiably determined with a stated
acceptable accuracy. For a hematology system, LoQ is determined based on the imprecision for low-
concentration samples. Thus, the concentration at which the CV% is smaller than the desired or claimed
imprecision for each parameter is the LoQ.

7.3.1. Samples
On each of three days of testing, four low-concentrate samples (WBC range 0.11–0,76 × 109/L, PLT range:
7–30 × 109/L) were tested five times (60 analyses).

7.3.2. Test procedure

To determine LoQ, four different samples with low concentration of PLT and WBC were analyzed with two
different reagent pairs. Five replicates of each sample were analyzed with each reagent pair. The mean,
SD, and CV% were calculated for each sample and the LoQ was determined as the lowest concentration
where the specification (WBC CV ≤ 15%, PLT CV ≤ 25%) was fulfilled.

7.3.3. Test results

LoQ results are shown in Table 7-3.

Table 7-3. LoQ of Swelab Lumi

Parameter WBC PLT

LoQ 0.18 × 109/L 20 × 109/L

10
8. Analytical specificity
Interfering substances can impact hematology parameter results.

8.1. Samples
It is recommended to design interference studies to mimic actual processes, testing increasing
concentrations of the interferent with the analyte of interest. The following interferences were studied:

• 
Triglycerides: 12 blood fat samples with different concentrations were selected for analysis of
interference with HGB measurements.

• 
Total Bilirubin: 12 jaundice samples with different concentrations were selected for analysis of
interference with HGB measurements.

• 
NRBC: 5 samples containing NRBC from patients with severe anemia or myelopathic anemia
were selected for analysis (the working principle of the reference analyzer for testing WBC is the
fluorescence scattering method, which will not interfere with NRBC).

• 
WBC turbidity: samples with WBC > 100 × 109/L were selected for analysis of interference with WBC
and HGB measurements.

8.2. Test procedure

Each interferent level was analyzed on the test and reference analyzers. Bias outside of the specification
limits was considered an interference.

8.3. Test results

The level of interference in percent was calculated as the relative difference between the test and
reference systems. The results show that high NRBC will result in falsely high WBC counts, whereas
triglyceride concentration of up to 15.39 mmol/L, total bilirubin concentration of up to 433.1 µmol/L, nor
high WBC will give interference.

9. WBC classification – clinical sensitivity and specificity

Sensitivity is the proportion of results from the test system that are positive (abnormal, outside established
reference interval) when the results from the reference method are positive (abnormal). Specificity is the
proportion of results from the test system that are negative (normal, within established reference interval)
when the results from the reference method are negative (normal). Sensitivity and specificity can take
values between 0% and 100%. True and false results can only be identified in relation to an independent
reference method.

9.1. Samples
Fresh blood samples (n = 200; 100 normal, 100 abnormal) were collected for use in the predictive value
performance study.

9.2. Test procedure

Each sample were analyzed on the test system and with the reference method (microscopy), and results
were collected for NEU, LYM, MON, EOS, and BAS. For the reference method, 400 cells from each blood
sample were analyzed. Two qualified testers analyzed 200 cells per blood smear. For the test system, all
samples were analyzed in duplicate.

Note! When the test result of the reference method is 0 and the test result of the analyzer is ≤ 1.0%, the test
conclusion is qualified. The test procedure is according to CLSI H20-A2.

11
9.3. Test results
The results are compiled in Table 9-1.

Table 9-1. WBC classification accuracy

Statistic Distribution Morphology Flagging

Efficiency 83.5% 97.5% 83.5%
Sensitivity 83.3% 100.0% 83.3%
Specificity 83.9% 97.5% 83.9%
Positive predictive value 92.0% 28.6% 92.0%
Negative predictive value 69.3% 100.0% 69.3%

10. Carryover
Carryover is a measure of the amount of analyte that is transferred from one sample measurement into
the following sample measurement.

10.1. Samples
Two whole blood samples from donors were processed to achieve high and low parameter levels.

10.2. Test procedure

Sample with the high-concentration (HC) parameter was analyzed three times in three analyzers.
Immediately after this, the sample with the low-concentration (LC) parameter was analyzed three times,
in the same three analyzers. Steps 1–2 were repeated four times, for a total number of analyses of 30 per
tested parameter and analyzer, that is, (3HC + 3LC) × 5.

10.3. Test results

The study outcome compiled in Table 10-1 shows that the carry-over performance of Swelab Lumi system
fulfills the claimed requirement according to specifications.

Table 10-1. Carryover in Swelab Lumi

WBC RBC HGB PLT HCT

Carryover (%), Device 1 0.0% 0.1% 0.4% 0.2% 0.1%
Carryover (%), Device 2 0.1% 0.2% 0.0% 0.1% 0.0%
Carryover (%), Device 3 0.0% 0.1% 0.4% 0.3% 0.1%
Carryover (%), specification 0.5% 0.5% 0.5% 1.0% 0.5%

12
11. Reference intervals
Demographic characteristics such as age, gender, and ethnicity can impact hematology parameters.
As stated in the Swelab Lumi User manual, the reference intervals in the default profile configuration are
indicative reference intervals and it is recommended for the laboratory to establish their own reference
intervals.

11.1. Samples
Fresh normal human whole blood samples from healthy male (n = 20) and female (n = 20) adult donors and
healthy children of different age groups (each: n = 20) were used to establish the reference intervals.

11.2. Test procedure

Reference intervals were selected according to literature:

• 
Adults: WST 405-2012 Reference interval for blood cell analysis issued by the Health Industry
Standards of the People’s Republic of China

• 
Children: WST 779-2021 Reference interval for blood cell analysis for children

Samples were analyzed and the number of results that met the reference interval was recorded. If out-
of-range samples were ≤ 2, the set reference interval was accepted. If out-of-range samples were 3, the
analysis was repeated with a new set of samples. Out-of-range samples of ≥ 4 were considered due to
biological or demographic differences, for which a new test procedure should be established.

11.3. Test results

Table 11-1 shows the default reference intervals of Swelab Lumi for different genders and age groups.

Table 11-1. Combined male and female reference intervals established on the Swelab Lumi system

MCHC
WBC

MON

MON

MCH
MCV
Parameter
HGB
RBC

HCT
EOS

EOS
NEU

NEU
BAS

BAS
LYM

LYM

PLT
Unit 109/L % % % % % 109/L 109/L 109/L 109/L 109/L 1012/L g/dL % fL pg g/dL 109/L
Adult male
Lower limit 3.5 40 20 3 0.4 0 1.8 1.1 0.1 0.02 0 4.3 130 40 82 27 316 125
Upper limit 9.5 75 50 10 8 1 6.3 3.2 0.6 0.52 0.06 5.8 175 50 100 34 354 350
Meeting rate 100% 100% 100% 100% 95% 100% 100% 100% 100% 100% 100% 100% 100% 90% 100% 100% 95% 100%
Requirement (≥) 90%
Adult female
Lower limit 3.5 40 20 3 0.4 0 1.8 1.1 0.1 0.02 0 3.8 115 35 82 27 316 125
Upper limit 9.5 75 50 10 8 1 6.3 3.2 0.6 0.52 0.06 5.1 150 45 100 34 354 350
Meeting rate 100% 100% 100% 100% 95% 100% 100% 100% 100% 95% 100% 100% 100% 100% 100% 100% 100% 95%
Requirement (≥) 90%
Children 28 d to 6 mo
Lower limit 5.6 7 34 3 0.8 0 0.6 3.2 0.25 0.06 0 3.5 99 29 73 24 305 203
Upper limit 14.5 51 81 18 11 1 7.1 10.7 1.89 1.22 0.14 5.6 196 57 105 37 361 653
Meeting rate 100% 100% 100% 95% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100% 100%
Requirement (≥) 90%
Children 6 mo to 1 yr
Lower limit 5 9 37 2 0.8 0 0.8 2.8 0.15 0.06 0 4.1 103 32 71 24 309 172
Upper limit 14.2 53 82 14 11 1 6.1 10 1.24 1.22 0.14 5.5 138 45 86 30 359 601
Meeting rate 100% 95% 95% 90% 90% 90% 100% 90% 100% 90% 100% 95% 100% 100% 100% 95% 100% 100%
Requirement (≥) 90%

13
Continuation of Table 11-1.

MCHC
WBC

MON

MON

MCH
MCV
Parameter

HGB
RBC

HCT
EOS

EOS
NEU

NEU
BAS

BAS
LYM

LYM

PLT
Unit 109/L % % % % % 109/L 109/L 109/L 109/L 109/L 1012/L g/dL % fL pg g/dL 109/L
Children 1 to 2 yr
Lower limit 5.5 13 35 2 0.5 0 0.9 2.7 0.2 0.04 0 4.1 104 32 71 24 309 191
Upper limit 13.6 54 76 14 9 1 5.5 9.1 1.14 0.74 0.1 5.5 143 43 86 30 359 516
Meeting rate 90% 90% 100% 100% 95% 100% 90% 95% 100% 100% 100% 100% 100% 100% 100% 90% 100% 95%
Requirement (≥) 90%
Children 2 to 6 yr
Lower limit 4.9 23 26 2 0.5 0 1.3 2 0.16 0.04 0 4.1 115 35 76 24 309 187
Upper limit 12.7 64 67 11 9 1 6.7 6.5 0.92 0.74 0.1 5.5 150 45 88 30 359 475
Meeting rate 100% 90% 90% 100% 95% 90% 90% 95% 100% 90% 90% 100% 100% 100% 100% 100% 100% 100%
Requirement (≥) 90%
Children 6 to 13 yr
Lower limit 4.6 32 22 2 0.5 0 1.7 1.7 0.15 0.04 0 4.3 121 37 77 26 309 177
Upper limit 11.9 71 57 11 9 1 7.4 4.7 0.86 0.74 0.1 5.7 158 47 92 34 359 446
Meeting rate 100% 90% 90% 90% 90% 95% 100% 90% 90% 90% 100% 100% 100% 100% 100% 100% 100% 100%
Requirement (≥) 90%
Male children 13 to 18 yr
Lower limit 4.1 37 17 2 0 0 1.8 1.2 0.14 0 0 4.5 129 39 80 25 310 150
Upper limit 11 77 54 11 9 1 8.3 3.8 0.74 0.68 0.07 5.9 172 51 100 34 355 407
Meeting rate 100% 100% 100% 100% 95% 90% 100% 95% 95% 95% 100% 95% 95% 95% 100% 100% 100% 100%
Requirement (≥) 90%
Female children 13 to 18 yr
Lower limit 4.1 37 17 2 0 0 1.8 1.2 0.14 0 0 4.1 114 36 80 25 310 150
Upper limit 11 77 54 11 9 1 8.3 3.8 0.74 0.68 0.07 5.3 154 47 100 34 355 407
Meeting rate 95% 95% 95% 100% 100% 90% 90% 100% 95% 100% 100% 95% 95% 90% 90% 100% 100% 100%
Requirement (≥) 90%

12. Analytical measuring ranges (linearity ranges)

Linearity is referred to as the ability of an instrument, given different samples with known and varied
concentrations of an analyte, to measure and present within the acceptance requirements.

12.1. Samples
Gradient dilutions of samples with high levels of WBC, RBC, and PLT were done to obtain at least 5
concentration levels. The highest concentration should be close to or slightly exceed the upper limit of the
linear range, and the lowest concentration should be close to or slightly below the lower limit of the linear
range.

12.2. Test procedure

Each concentration level was analyzed three times, and the mean was calculated for each concentration
level. From the plotted mean (y) against the theoretical concentration (x), the absolute or relative
deviation between the mean and the theoretical value was calculated.

14
12.3. Test results
Test results shown in Figure 12-1 and the analytical measurement ranges compiled in Table 12-1 show that
the linearity of Swelab Lumi is within the claimed ranges.

(A) (B)
120 350
WBC Linearity WBC Linearity
y = -0.295 + 2.968 x y = -9.064 + 3.236 x
100 n = 10 n=6
300
r = 1.00; P < 0.001 r = 1.00; P < 0.001
80
250
WBC_Mean

WBC_Mean
60
200
40

150
20

0 100
0 10 20 30 40 40 50 60 70 80 90 100
Concentration (%) Concentration (%)

(C) (D)
9 300
RBC Linearity HGB Linearity
8 y = 0.0379 + 0.0871 x y = 1.680 + 2.831 x
n=9 250 n=9
7
r = 1.00; P < 0.001 r = 1.00; P < 0.001
6 200
HGB_Mean
RBC_Mean

5
150
4
3 100

2
50
1

0 0
0 20 40 60 80 100 0 20 40 60 80 100
Concentration (%) Concentration (%)

(E) (F)
3500 100
PLT Linearity HCT Linearity
3000 y = -5.036 + 34.315 x y = 0.690 + 0.892 x
n = 14 80 n=9
r = 1.00; P < 0.001 r = 1.00; P < 0.001
2500

60
HCT_Mean

2000
PLT_Mean

1500
40

1000
20
500

0 0
0 20 40 60 80 100 0 20 40 60 80 100
Concentration (%) Concentration (%)

Fig 12-1. Linearity test results for (A) WBC ([0.00–100.00] × 109/L), (B) WBC ([100.01–300.00] × 109/L), (C) RBC, (D) HGB,
(E) PLT, and (F) HCT.

15
Table 12-1. Linearity results obtained from Swelab Lumi

Parameter Theoretical conc. Mean value ranges Difference r2 Linearity range established
ranges (whichever is greater)
WBC (0.3125–40)% (0.92–119.48) × 109/L ± 3.0% × 109/L or ± 5.0% 1 (0.00–100.00) × 109/L
(40–100)% (119.48–313.54) × 10 /L
9
± 1 0% 1 (100.01–300.00) × 109/L
RBC (0.3125–100)% (0.29–8.63) × 1012/L ± 0.05 × 1012/L or ± 5% 1 (0.00–8.50) × 1012/L
HGB (0.3125–100)% (9.00–281.33) g/L ± 2 g/L or ± 2% 1 (0–250) g/L
PLT (0.3125–100)% (6–3376) × 10 /L
9
± 10 × 10 /L or ± 8%
9
1 (0–1000) × 109/L
RBC ≤ 7.0 RBC ≤ 7.0
± 12% 1 (1001–3000) × 109/L
RBC ≤ 7.0
HCT (0.3125–100)% (3.10–88.27)% ± 2% (HCT value) 1 0.0%–67.0%
or ±3% (error percentage)

13. Mode-to-mode comparability

Swelab Lumi has the following analysis modes:
1. CBC
2. CBC + DIFF

To demonstrate that Swelab Lumi performs equally on samples tested in the different analysis modes, a
mode-to-mode comparability study was conducted and the results from each parameter recorded, and
the correlation coefficient (r) was calculated by statistical regression equation: y = ax + b.

The following samples were excluded:

• 
Samples including microcytes
• 
Samples including nucleated red blood cells
• 
Samples including immature cells

Obtained results were comparable and approved for both analysis modes.

boule.com
Swelab is a trademark of Boule Medical AB. SYSMEX is a trademark of Sysmex Corp.
© 2023 Boule Diagnostics AB
TR 41009, 41672
Boule Diagnostics AB, Domnarvsgatan 4, SE-163 53 Spånga, Sweden
PSS44278-1 03/2023