1) The study investigated the significance of blood cell levels in early detection of lung cancer by analyzing data from 115 lung cancer patients and 120 healthy individuals. 2) Neural network analysis revealed that early lung cancer detection depended most on lymphocyte levels, followed by neutrophil levels, monocytes, and other blood cells. 3) The neural networks achieved 100% accurate detection of early lung cancer based on blood cell levels, demonstrating their potential for precise early detection.

1. Precise Early Detection of Lung
Cancer and Blood Cell Circuit
Oleg Kshivets, MD, PhD Surgery Department, Roshal Hospital, Roshal, Moscow, Russia P2.11
Objective: Significance of blood cell circuit in terms of early detection of lung cancer (LC)
was investigated.
Methods: In trial (1987-2018) consecutive cases after surgery, monitored 115 LC patients
(LCP) (m=100, f=15; lobectomies=115) with pathologic stage IA (tumor size=1.86±0.30 cm;
squamous=51, adenocarcinoma=59, large cell=5; T1N0M0=115; G1=39, G2=42, G3=34, 5-year
survival=100%) and 120 healthy donors (HD) (m=69, f=51) were reviewed. Variables selected for
study were input levels of blood cell circuit, sex, age, TNMG. Differences between groups were
evaluated using discriminant analysis, clustering, nonlinear estimation, structural equation
modeling, Monte Carlo, bootstrap simulation and neural networks computing.
Results: It was revealed that early detection of LC from HD (n=235) significantly depended
on: leucocytes (abs, total), segmented neutrophils (%, abs, total), lymphocytes (%), monocytes
(abs, total) (P=0.012-0.000). Neural networks computing, genetic algorithm selection and
bootstrap simulation revealed relationships of early detection of LC and lymphocytes (rank=1),
stick neutrophils (rank=2), monocytes (3), segmented neutrophils (4), leucocytes (5),
eosinophils (6). Correct detection of early LCP was 100% by neural networks computing
(error=0.000; area under ROC curve=1.0).
Conclusion: Early detection of LC from HD significantly depended on blood cell circuit.
Blood Factors Mean -
Norm
Mean -
Cancer
t-value p Std.Dev. -
Norm
Std.Dev. -
Cancer
Leucocytes 4.9650 5.8704 -3.49567 0.000566 1.58240 2.33204
Stick Neutrophils 1.0583 1.0957 -0.21389 0.830823 1.32396 1.35059
Segmented Neutrophils 55.8583 61.4609 -3.93139 0.000111 10.68990 11.15622
Eosinophils 3.2250 3.0562 0.41205 0.680682 3.66556 2.47445
Monocytes 3.1167 3.4609 -0.99151 0.322467 2.38406 2.92085
Lymphocytes 36.8250 30.9478 4.27742 0.000028 10.32714 10.73594
Eosinophils abs 0.1632 0.1790 -0.59403 0.553066 0.23783 0.16201
Stick Neutrophils abs 0.0591 0.0709 -0.86079 0.390238 0.08058 0.12631
Segmented Neutrophils abs 2.8110 3.6048 -4.48441 0.000011 1.17819 1.52038
Lymphocytes abs 1.7805 1.8052 -0.19326 0.846926 0.68395 1.21320
Monocytes abs 0.1532 0.2119 -2.54010 0.011733 0.13311 0.21328
Eosinophils tot 0.7865 0.8638 -0.59394 0.553125 1.17369 0.77206
Stick Neutrophils tot 0.2848 0.3509 -0.96799 0.334052 0.40818 0.62059
Segmented Neutrophils tot 13.4859 17.7378 -4.58500 0.000007 6.00216 8.09992
Lymphocytes tot 8.5241 8.8008 -0.43961 0.660628 3.38285 5.96624
Monocytes tot 0.7315 1.0456 -2.76601 0.006130 0.65846 1.04679
Leucocytes tot 23.8069 28.7939 -3.66118 0.000311 8.30366 12.27662
Blood Factors
N=235
Discriminant Function Analysis Summary (EDLC_N1) No. of
vars in model: 17; Grouping: CA--NORM (2 grps) Wilks'
Lambda: .83538 approx. F (17,217)=2.5155 p< .0012
Wilks'
Lambda
Partial
Lambda
F-
remove
(1,217)
p-value Toler.
1-Toler.
(R-Sqr.)
Leucocytes 0.835539 0.999805 0.042398 0.837055 0.000005 0.999995
Stick Neutrophils 0.847328 0.985894 3.104805 0.079470 0.031455 0.968545
Segmented Neutrophils 0.848397 0.984652 3.382399 0.067262 0.000486 0.999514
Eosinophils 0.849547 0.983319 3.681257 0.056340 0.005576 0.994424
Monocytes 0.852830 0.979533 4.534138 0.034352 0.007193 0.992807
Lymphocytes 0.850429 0.982299 3.910244 0.049258 0.000521 0.999479
Eosinophils abs 0.835376 1.000000 0.000103 0.991893 0.000478 0.999522
Stick Neutrophils abs 0.835500 0.999851 0.032310 0.857518 0.001454 0.998546
Segmented Neutrophils abs 0.835594 0.999738 0.056859 0.811757 0.000011 0.999989
Lymphocytes abs 0.835476 0.999880 0.026101 0.871804 0.000021 0.999979
Monocytes abs 0.835773 0.999524 0.103267 0.748253 0.000577 0.999423
Eosinophils tot 0.837467 0.997502 0.543392 0.461827 0.000609 0.999391
Stick Neutrophils tot 0.835724 0.999583 0.090514 0.763813 0.001723 0.998277
Segmented Neutrophils tot 0.839000 0.995681 0.941352 0.333011 0.000012 0.999988
Lymphocytes tot 0.838435 0.996351 0.794781 0.373646 0.000026 0.999974
Monocytes tot 0.841001 0.993311 1.461255 0.228046 0.000718 0.999282
Leucocytes tot 0.838741 0.995987 0.874275 0.350814 0.000006 0.999994
N=235
Model: Logistic regression (logit) N of 0's: 120 1's: 115 Dep. var: Ca-Norm Loss: Max likelihood Final loss: 140.17256193 Chi2 (17)=45.328 p=.00022
Modeled probability that ST = 1.
Const.B0 L ST S E M LYM Eabs STabs Sabs LYMabs Mabs Etot STtot Stot LYMtot Mtot Ltot
Estimate 137.845 6.0578 -1.27 -1.3505 -1.4148 -1.7565 -1.4314 1.3690 10.7 -7.347 -4.1421 -12.05 4.7529 1.586 6.4389 5.8372 9.194 -6.1893
Odds ratio
(unit ch)
427.44 0.2813 0.2591 0.2430 0.1727 0.2390 3.9315 42852 0.0006 0.01589 0.000 115.92 4.882 625.739 342.809 9842.75 0.0021
Odds ratio
(range)
0.0005 0.0000 0.0000 0.0000 0.0000 19.136 138511 0.000 0.0000 0.000 5563.08 0.0000
Dependent
Variable
Test of SS Whole Model vs. SS Residual
Multiple
R
Multiple
R2
Adjusted
R2
SS
Model
df
Model
MS
Model
SS
Residual
df
Residual
MS
Residual
F p
CA_NORM 0.405739 0.164624 0.099180 9.667307 17 0.568665 49.05610 217 0.226065 2.515494 0.001154
Neural Networks: n=235
Baseline Error=0.000;
Area under ROC Curve=1.000;
Correct Classification Rate=100%
Rank Sensitivity
Lymphocytes 1 3076
Stick Neutrophils 2 2323
Monocytes 3 2216
Segmented Neutrophils 4 2018
Leucocytes 5 1917
Eosinophils 6 1771
Bootstrap Simulation
Significant Factors
(Number of Samples=3333)
Rank Kendall’
Tau-A
P<
Segmented Neutrophils tot 1 -0.191 0.001
Segmented Neutrophils abs 2 -0.172 0.01
Lymphocytes 3 0.162 0.01
Segmented Neutrophils 4 -0.149 0.01
Leucocytes tot 5 -0.140 0.01
Leucocytes 6 -0.125 0.01