There is neither a single clinical, microbiological, histopathological or genetic test, nor combinations of them, to discriminate aggressive periodontitis (AgP) from chronic periodontitis (CP) patients. We aimed to estimate probability density functions of clinical and immunologic datasets derived from periodontitis patients and construct artificial neural networks (ANNs) to correctly classify patients into AgP or CP class. The fit of probability distributions on the datasets was tested by the Akaike information criterion (AIC). ANNs were trained by cross entropy (CE) values estimated between probabilities of showing certain levels of immunologic parameters and a reference mode probability proposed by kernel density estimation (KDE). The weight decay regularization parameter of the ANNs was determined by 10-fold cross-validation. Possible evidence for 2 clusters of patients on cross-sectional and longitudinal bone loss measurements were revealed by KDE. Two to 7 clusters were shown on datasets of CD4/CD8 ratio, CD3, monocyte, eosinophil, neutrophil and lymphocyte counts, IL-1, IL-2, IL-4, INF-γ and TNF-α level from monocytes, antibody levels against A. actinomycetemcomitans (A.a.) and P.gingivalis (P.g.). ANNs gave 90%–98% accuracy in classifying patients into either AgP or CP. The best overall prediction was given by an ANN with CE of monocyte, eosinophil, neutrophil counts and CD4/CD8 ratio as inputs. ANNs can be powerful in classifying periodontitis patients into AgP or CP, when fed by CE values based on KDE. Therefore ANNs can be employed for accurate diagnosis of AgP or CP by using relatively simple and conveniently obtained parameters, like leukocyte counts in peripheral blood. This will allow clinicians to better adapt specific treatment protocols for their AgP and CP patients.
References
[1]
Kinane DF, Berglund T, Lindhe J (2008) Pathogenesis of periodontitis. In: Lindhe J, Karring T, Lang NP, editors. Clinical Periodontology and Implant Dentistry. Blackwell Munksgaard: Oxford. pp. 285–306.
[2]
Eke PI, Dye BA, Wei L, Thornton-Evans GO, Genco RJ (2012) Prevalence of periodontitis in adults in the United States: 2009 and 2010. J Dent Res 91: 914–920. doi: 10.1177/0022034512457373
[3]
Laine ML, Crielaard W, Loos BG (2012) Genetic susceptibility to periodontitis. Periodontol 2000 58: 37–68. doi: 10.1111/j.1600-0757.2011.00415.x
[4]
Nicolis G, Nicolis C (2012) Foundations of complex systems: emergence, information and prediction. World Scientific: Singapore. 12 p.
[5]
Armitage GC (1999) Developement of a classification system for periodontal diseases and conditions. Ann Periodontol 4: 1–6.
[6]
Armitage GC (2013) Learned and unlearned concepts in periodontal diagnostics: a 50-year perspective. Periodontol 2000 62: 20–36. doi: 10.1111/prd.12006
[7]
Tonetti MS, Mombelli A (2008) Aggressive periodontitis. In: Lindhe J, Karring T, Lang NP, editors. Clinical Periodontology and Implant Dentistry. Blackwell Munksgaard: Oxford. pp. 428–458.
[8]
Armitage GC, Cullinan MP (2010) Comparison of the clinical features of chronic and aggressive periodontitis. Periodontol 2000 53: 12–27. doi: 10.1111/j.1600-0757.2010.00353.x
[9]
Papantonopoulos G, Takahashi K, Bountis T, Loos BG (2013a) Using cellular automata experiments to model periodontitis: A first step towards understanding the nonlinear dynamics of the disease. Int J Bifurcation Chaos 23: 1350056. doi: 10.1142/s0218127413500569
Silverman BW (1986) Density estimation for statistics and data analysis. Chapman and Hall: London. 6 p.
[12]
Simonoff JS (1996) Smoothing methods in statistics. New York: Springer-Verlag: New York. 13 p.
[13]
Sheather SJ (2004) Density estimation. Stat Science 19: 588–597. doi: 10.1214/088342304000000297
[14]
Abdi H (2003) Neural networks. In: Lewis-Beck M, Bryman A, Futing T, editors. Encyclopaedia of Social Sciences Research Methods. Sage: Thousand Oaks (CA). pp. 725–729.
[15]
Papantonopoulos G (2004) Effect of periodontal therapy in smokers and nonsmokers with advanced periodontal disease: Results after maintenance therapy for a minimum of 5 years. J Periodontol 75: 839–843. doi: 10.1902/jop.2004.75.6.839
[16]
Loos BG, Craandijk J, Hoek FJ, Wertheim-van Dillen PME, van der Velden U (2000) Elevation of systemic markers related to cardiovascular diseases in the peripheral blood of periodontitis patients. J Periodontol 71: 1528–1534. doi: 10.1902/jop.2000.71.10.1528
[17]
Loos BG, Roos MT, Schellekens PT, van der Velden U, Miedema F (2004) Lymphocyte number and function in relation to periodontitis and smoking. J Periodontol 75: 557–564. doi: 10.1902/jop.2004.75.4.557
[18]
Graswinckel JEM, van der Velden U, van Winkelhoff AJ, Hoek FJ, Loos BG (2004) Plasma antibody levels in periodontitis patients and controls. J Clin Periodontol 31: 562–568. doi: 10.1111/j.1600-051x.2004.00522.x
[19]
Takahashi K, Ohyama H, Kitanaka M, Sawa T, Mineshiba J, et al. (2001) Heterogeneity of host immunological risk factors in patients with aggressive periodontitis. J Periodontol 72: 425–437. doi: 10.1902/jop.2001.72.4.425
[20]
Kinane DF (2001) Periodontal disease in children and adolescents: introduction and classification. Periodontol 2000 26: 7–15. doi: 10.1034/j.1600-0757.2001.2260101.x
[21]
Hall P, Kang KH (2005) Bandwidth choice for nonparametric classification. Ann Statist 33: 284–306. doi: 10.1214/009053604000000959
[22]
Vose D (2008) Risk analysis: A quantitative guide. West Sussex: John Wiley & Sons. 115 p.
[23]
Sakamoto Y, Ishikuro M, Kitaqawa G (1986) Akaike information criterion statistics. Springer-Verlag: New York. 56 p.
[24]
Myung J (2003) Tutorial on maximum likelihood estimation. J Math Psychol 47: 90–100. doi: 10.1016/s0022-2496(02)00028-7
[25]
Burnham KP, Anderson DR (2004) Multimodel inference: understanding AIC and BIC in model selection. Sociol Methods Res 33: 261–304. doi: 10.1177/0049124104268644
[26]
de Boer PT, Kroese DP, Mannor S, Rubinstein RY (2005) A tutorial on the cross-entropy method. Ann Oper Res 134: 18–67. doi: 10.1007/s10479-005-5724-z
[27]
Neal R, Zhang J (2006) High dimensional classification with bayesian neural networks and dirichlet diffusion trees. In: Guyon I, Gunn S, Nikravesh M, Zadeh L, editors. Feature Extraction, Foundations and Applications. Springer: New York. pp. 265–296.
[28]
Hastie T, Tibshirani R, Friedman J (2009) The elements of statistical learning: Data mining, inference, and prediction. Springer: New York. 389 p.
[29]
Zhang GP (2000) Neural networks for classification: A survey. IEEE Trans Syst Man Cybern 30: 451–462. doi: 10.1109/5326.897072
[30]
Mazurowski MA, Habas PA, Zurada JM, Lo JY, Baker JA, et al. (2008) Training neural network classifiers for medical decision making; The effects of imbalanced datasets on classification performance. Neural Netw 21: 427–436. doi: 10.1016/j.neunet.2007.12.031
[31]
Shi H-Y, Lee K-T, Lee H-H, Ho W-H, Sun D-P, et al. (2012) Comparison of artificial neural network and logistic regression models for predicting in-hospital mortality after primary liver cancer surgery. PLOS ONE 7 (4) e35781. doi: 10.1371/journal.pone.0035781
[32]
Offenbacher S, Barros SP, Singer RE, Moss K, Williams RC, et al. (2007) Periodontal disease at the biofilm-gingival interface. J Periodontol 78: 1911–1925. doi: 10.1902/jop.2007.060465
[33]
Teles RP, Gursky LC, Faveri M, Rosa EA, Teles FRF, et al. (2010) Relationships between subgingival microbiota and GCF biomarkers in generalized aggressive periodontitis. J Clin Periodontol 37: 313–323. doi: 10.1111/j.1600-051x.2010.01534.x
[34]
Razzouk S, Termechi O (2013) Host genome, epigenome, and oral microbiome interactions: towards personalized periodontal therapy. J Periodontol 84: 1266–1271. doi: 10.1902/jop.2012.120531
[35]
Loos BG, Tjoa S (2005) Host-derived diagnostic markers for periodontitis: do they exist in gingival crevicular fluid? Periodontol 2000 39: 53–72. doi: 10.1111/j.1600-0757.2005.00129.x
[36]
Bartold PM, Van Dyke E (2013) Periodontitis: a host-mediated disruption of microbial homeostasis. Unlearning learned concepts. Periodontol 2000 62: 203–217. doi: 10.1111/j.1600-0757.2012.00450.x
[37]
Guyon I, Elisseeff A (2003) An introduction to variable and feature selection. J Mach Learn Res 3: 1157–1182.
[38]
Van der Velden U (2005) Purpose and problems of periodontal disease classification. Periodontol 2000 39: 13–21. doi: 10.1111/j.1600-0757.2005.00127.x
[39]
Kebschull M, Guarnieri P, Demmer RT, Boulesteix AL, Pavlidis P, et al. (2013) Molecular Differences between Chronic and Aggressive Periodontitis. J Dent Res 92: 1081–1088. doi: 10.1177/0022034513506011
