Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

Kshivets chicago2016

6,327 views

Published on

Artificial Intelligence, System Analysis and Simulation Modeling in Optimization of Management for Lung Cancer Patients

Published in: Health & Medicine

Kshivets chicago2016

  1. 1. Artificial Intelligence, System Analysis and Simulation Modeling in Optimization of Management for Lung Cancer Patients Oleg Kshivets, MD, PhD Surgery Department, Khimki Clinic N1, Moscow, Russia PS-01.40 OBJECTIVE: Search of best management for non-small lung cancer (LC) patients (LCP) (T1-4N0-2M0) was realized. METHODS: We analyzed data of 672 consecutive LCP (age=57.5±8.3 years; tumor size=4.4±2.4 cm) radically operated and monitored in 1985-2016 (m=581, f=91; lobectomies=430, pneumonectomies=242, combined procedures with resection of trachea, carina, atrium, aorta, VCS, vena azygos, pericardium, liver, diaphragm, ribs, esophagus=184; only surgery=530, adjuvant chemoimmunoradiotherapy-AT=142: CAV/gemzar + cisplatin + thymalin/taktivin + radiotherapy 45-50Gy; T1=239, T2=249, T3=129, T4=55; N0=425, N1=130, N2=117, M0=672; G1=168, G2=201, G3=303; squamous=380, adenocarcinoma=247, large cell=45; early LC=134, invasive LC=538. Multivariate Cox modeling, clustering, SEPATH, Monte Carlo, bootstrap and neural networks computing were used to determine any significant dependence. RESULTS: Overall life span (LS) was 2104.2±1685.2 days and cumulative 5-year survival (5YS) reached 69.7%, 10 years – 61.9%, 20 years – 42.5%. 417 LCP lived more than 5 years (LS=3044.7±1472.2 days), 109 LCP – more than 10 years (LS=5048±1471.6 days). 194 LCP died because of LC (LS=560.7±372.9 days). AT significantly improved 5YS (65.3% vs. 34.3%) (P=0.00001 by log-rank test) only for LCP with N1-2. Cox modeling displayed (Chi2=292.44, df=13, P=0.000) that 5YS of LCP significantly depended on: phase transition (PT)“early-invasive LC” in terms of synergetics, PT N0-N12, histology, G, blood cell subpopulations, cell ratio factors (ratio between blood cells subpopulations and cancer cells-CC), prothrombin index, heparin tolerance, recalcification time, glucose, AT (P=0.000-0.033). Neural networks, genetic algorithm selection and bootstrap simulation revealed relationships between 5YS and PT“early-invasive LC” (rank=1), PT N0-N12 (rank=2), AT (3), segmented neutrophils (4), lymphocytes (5), prothrombin index (6),healthy cells/CC (7), T1-4 (8), tumor size (9), thrombocytes/CC (10), erythrocytes/CC (11),lymphocytes/CC (12). Correct prediction of 5YS was 100% by neural networks computing (error=0.000; area under ROC curve=1.0). CONCLUSIONS: Optimal management for LCP are: 1) screening and early detection of LC; 2) availability of experienced thoracic surgeons because of complexity of radical procedures; 3) aggressive en block surgery and adequate lymph node dissection for completeness; 4) precise prediction; 5) adjuvant chemoimmunoradiotherapy for LCP with unfavorable prognosis. Parameter Estimates:Chi2=292.44, df=13, P=0.0000, n=672 Cox Regression Parameter Estimate Standard Error Chi- square P value 95% Lower CL 95% Upper CL Hazard Ratio 95% HazardRatio Lower CL 95% HazardRatio Upper CL Histology 0.29021 0.087856 10.91150 0.000956 0.11802 0.462408 1.336711 1.125263 1.587893 G1-3 0.32570 0.089528 13.23471 0.000275 0.15023 0.501169 1.384997 1.162098 1.650649 Phase Transition Early-Invasive Cancer -1.32520 0.393120 11.36353 0.000749 -2.09570 -0.554700 0.265749 0.122984 0.574244 Adjuvant Chemoimmunoradiotherapy -1.01228 0.201036 25.35421 0.000000 -1.40630 -0.618253 0.363391 0.245048 0.538885 Phase Transition N0--N12 1.06470 0.147841 51.86362 0.000000 0.77493 1.354459 2.899959 2.170449 3.874666 Glucose -0.29033 0.081154 12.79817 0.000347 -0.44938 -0.131266 0.748020 0.638020 0.876984 Prothrombin Index 0.03002 0.006828 19.32862 0.000011 0.01664 0.043400 1.030473 1.016775 1.044356 Recalcification Time -0.00585 0.001771 10.92016 0.000951 -0.00932 -0.002381 0.994166 0.990722 0.997622 Heparin Tolerance 0.00366 0.000667 30.07778 0.000000 0.00235 0.004965 1.003665 1.002353 1.004977 Eosinophils/Cancer Cells -0.99120 0.425130 5.43595 0.019726 -1.82443 -0.157957 0.371133 0.161309 0.853887 Lymphocytes/Cancer Cells -0.17877 0.067403 7.03422 0.007997 -0.31088 -0.046660 0.836300 0.732805 0.954412 Erythrocytes (tot) -0.03803 0.017819 4.55381 0.032846 -0.07295 -0.003101 0.962688 0.929647 0.996904 Thrombocytes (tot) 0.00050 0.000191 6.81611 0.009034 0.00012 0.000872 1.000498 1.000124 1.000872 Neural Networks: n=611 Baseline Error=0.000; Area under ROC Curve=1.000; Correct Classification Rate=100% Rank Sensitivity PT Early---Invasive Cancer 1 9516 PT N0---N12 2 9479 Adjuvant Chemoimmunoradiotherapy 3 6705 Segmented Neutrophils 4 5951 Lymphocytes 5 5894 Prothrombin Index 6 5126 Healthy Cells/Cancer Cells 7 5034 T1-4 8 4255 Tumor Size 9 3525 Thrombocytes/Cancer Cells 10 2840 Erythrocytes/Cancer Cells 11 2740 Lymphocytes/Cancer Cells 12 2170 Bootstrap Simulation Rank Kendall’Tau-A P< Lymphocytes/Cancer Cells 1 -0.227 0.000 Healthy Cells/Cancer Cells 2 -0.223 0.000 Erythrocytes/Cancer Cells 3 -0.221 0.000 PT N0---N12 4 0.197 0.000 Thrombocytes/Cancer Cells 5 -0.197 0.000 Leucocytes/Cancer Cells 6 -0.188 0.000 Prothrombin Index 7 0.163 0.000 Eosinophils/Cancer Cells 8 -0.161 0.000 Monocytes/Cancer Cells 9 -0.144 0.000 Tumor Size 10 0.143 0.000 Seg.Neutrophils/Cancer Cells 11 -0.134 0.000 PT Early---Invasive Cancer 12 -0.131 0.000 T1-4 13 -0.124 0.000 Lymphocytes (tot) 14 -0.108 0.000 Erythrocytes (tot) 15 -0.117 0.000 Segmented Neutrophils (%) 16 0.109 0.000 Weight 17 -0.105 0.000 Eosinophils (tot) 18 -0.085 0.01 Lymphocytes (%) 19 -0.086 0.01 G1-3 20 0.083 0.01 ESS 21 0.082 0.01 Lymphocytes (abs) 22 -0.082 0.01 Heparin Tolerance 23 0.078 0.01 Stick Neutrophils/Cancer Cells 24 -0.077 0.01 Glucose 25 -0.072 0.01 Only Surgery 26 -0.065 0.05 Eosinophils (%) 27 -0.063 0.05 Eosinophils (abs) 28 -0.059 0.05 0 5 10 15 20 25 30 35 Lymphocytes/Cancer Cells 11.11.21.31.41.51.61.71.81.9 PT Early--Invasive Cancer 1 1 1.1 1.1 1.2 1.2 1.3 1.3 1.4 1.4 1.5 1.5 1.6 1.6 1.7 1.7 1.8 1.8 1.9 1.9 2 2 5-YearSurvival 5-YearSurvival P=0.0000 z=a+blnx+cy+d(lnx)^2+ey^2+fylnx+g(lnx)^3+hy^3+iy^2lnx+jy(lnx)^2 r^2=0.15389193 DF Adj r^2=0.13979013 FitStdErr=0.43174167 Fstat=12.145684 a=1.6730247 b=0.10714131 c=-1.559062 d=-0.054738649 e=2.0848282 f=-0.42449195 g=0.0015228759 h=-0.73551088 i=0.18814164 j=0.022282855 0 5 10 15 20 25 30 35 Lymphocytes/Cancer Cells 00.10.20.30.40.50.60.70.80.9 PT N0---N12 1 1 1.1 1.1 1.2 1.2 1.3 1.3 1.4 1.4 1.5 1.5 1.6 1.6 1.7 1.7 1.8 1.8 1.9 1.9 2 2 5-YearSurvival 5-YearSurvival P=0.0000 z=a+b/x+c/x^2+d/x^3+e/x^4+f/x^5+gy r^2=0.24088672 DF Adj r^2=0.23207446 FitStdErr=0.40792762 Fstat=31.944196 a=0.85682428 b=1.0401408 c=-1.0202475 d=0.47891711 e=-0.098874149 f=0.0071729116 g=0.37035265

×