A functional supervised learning approach to the study of blood pressure data
Corresponding Author
Georgios I. Papayiannis
Department of Statistics, Athens University of Economics & Business, Athens, Greece
Sector of Mathematics, Hellenic Naval Academy, Piraeus, Greece
Stochastic Modeling and Applications Laboratory, Athens University of Economics & Business, Athens, Greece
Correspondence
Georgios I. Papayiannis, Patision Str. 76, Athens 10434, Greece.
Email: [email protected]
Search for more papers by this authorEmmanuel A. Giakoumakis
Department of Informatics, Athens University of Economics & Business, Athens, Greece
Search for more papers by this authorEfstathios D. Manios
Faculty of Medicine, National & Kapodistrian University of Athens, Athens, Greece
Search for more papers by this authorSpyros D. Moulopoulos
Faculty of Medicine, National & Kapodistrian University of Athens, Athens, Greece
Search for more papers by this authorKimon S. Stamatelopoulos
Faculty of Medicine, National & Kapodistrian University of Athens, Athens, Greece
Search for more papers by this authorSavvas T. Toumanidis
Faculty of Medicine, National & Kapodistrian University of Athens, Athens, Greece
Search for more papers by this authorNikolaos A. Zakopoulos
Faculty of Medicine, National & Kapodistrian University of Athens, Athens, Greece
Search for more papers by this authorAthanasios N. Yannacopoulos
Department of Statistics, Athens University of Economics & Business, Athens, Greece
Stochastic Modeling and Applications Laboratory, Athens University of Economics & Business, Athens, Greece
Search for more papers by this authorCorresponding Author
Georgios I. Papayiannis
Department of Statistics, Athens University of Economics & Business, Athens, Greece
Sector of Mathematics, Hellenic Naval Academy, Piraeus, Greece
Stochastic Modeling and Applications Laboratory, Athens University of Economics & Business, Athens, Greece
Correspondence
Georgios I. Papayiannis, Patision Str. 76, Athens 10434, Greece.
Email: [email protected]
Search for more papers by this authorEmmanuel A. Giakoumakis
Department of Informatics, Athens University of Economics & Business, Athens, Greece
Search for more papers by this authorEfstathios D. Manios
Faculty of Medicine, National & Kapodistrian University of Athens, Athens, Greece
Search for more papers by this authorSpyros D. Moulopoulos
Faculty of Medicine, National & Kapodistrian University of Athens, Athens, Greece
Search for more papers by this authorKimon S. Stamatelopoulos
Faculty of Medicine, National & Kapodistrian University of Athens, Athens, Greece
Search for more papers by this authorSavvas T. Toumanidis
Faculty of Medicine, National & Kapodistrian University of Athens, Athens, Greece
Search for more papers by this authorNikolaos A. Zakopoulos
Faculty of Medicine, National & Kapodistrian University of Athens, Athens, Greece
Search for more papers by this authorAthanasios N. Yannacopoulos
Department of Statistics, Athens University of Economics & Business, Athens, Greece
Stochastic Modeling and Applications Laboratory, Athens University of Economics & Business, Athens, Greece
Search for more papers by this authorAbstract
In this work, a functional supervised learning scheme is proposed for the classification of subjects into normotensive and hypertensive groups, using solely the 24-hour blood pressure data, relying on the concepts of Fréchet mean and Fréchet variance for appropriate deformable functional models for the blood pressure data. The schemes are trained on real clinical data, and their performance was assessed and found to be very satisfactory.
REFERENCES
- 1Pickering TG, Shimbo D, Haas D. Ambulatory blood-pressure monitoring. N Engl J Med. 2006; 354(22): 2368-2374.
- 2Kikuya M, Hansen TW, Thijs L, et al. Diagnostic thresholds for ambulatory blood pressure monitoring based on 10-year cardiovascular risk. Circulation. 2007; 115(16): 2145-2152.
- 3O'Brien E, Parati G, Stergiou G, et al. European society of hypertension position paper on ambulatory blood pressure monitoring. J Hypertens. 2013; 31(9): 1731-1768.
- 4Dolan E, Stanton A, Thijs L, et al. Superiority of ambulatory over clinic blood pressure measurement in predicting mortality the dublin outcome study. Hypertension. 2005; 46(1): 156-161.
- 5Ohkubo T, Hozawa A, Yamaguchi J, et al. Prognostic significance of the nocturnal decline in blood pressure in individuals with and without high 24-h blood pressure: the ohasama study. J Hypertens. 2002; 20(11): 2183-2189.
- 6Parati G, Valentini M, Metoki H, et al. Prognostic significance for stroke of a morning pressor surge and a nocturnal blood pressure decline: the ohasama study. commentary. Hypertension. 2006; 47(2): 149-154.
- 7Mancia G, Bombelli M, Facchetti R, et al. Long-term prognostic value of blood pressure variability in the general population results of the pressioni arteriose monitorate e loro associazioni study. Hypertension. 2007; 49(6): 1265-1270.
- 8Portaluppi F, Vergnani L, Manfredini R, et al. Endocrine mechanisms of blood pressure rhythms. Ann N Y Acad Sci. 1996; 783(1): 113-131.
- 9Kawano Y, Tochikubo O, Minamisawa K, et al. Circadian variation of haemodynamics in patients with essential hypertension: comparison between early morning and evening. J Hypertens. 1994; 12(12): 1405-1412.
- 10Portaluppi F, Smolensky MH. Circadian rhythmic and environmental determinants of 24-hour blood pressure regulation in normal and hypertensive conditions. Blood Pressure Monitoring in Cardiovascular Medicine and Therapeutics. New York: Springer; 2007: 135-158.
10.1007/978-1-59259-978-3_6 Google Scholar
- 11Sørensen H, Goldsmith J, Sangalli LM. An introduction with medical applications to functional data analysis. Stat Med. 2013; 32(30): 5222-5240.
- 12Wand MP, Jones CM. Kernel Smoothing. Boca Raton, Florida: CRC Press; 1994.
10.1201/b14876 Google Scholar
- 13Ramsay JO, Silverman BW. Functional Data Analysis. New York: Springer-Verlag; 2005.
10.1002/0470013192.bsa239 Google Scholar
- 14Ramsay JO, Li X. Curve registration. J R Stat Soc Series B (Statistical Methodology). 1998; 60(2): 351-363.
10.1111/1467-9868.00129 Google Scholar
- 15Gervini D, Gasser T. Self-modelling warping functions. J R Stat Soc Series B (Statistical Methodology). 2004; 66(4): 959-971.
10.1111/j.1467-9868.2004.B5582.x Google Scholar
- 16Marron JS, Ramsay JO, Sangalli LM, et al. Functional data analysis of amplitude and phase variation. Statist Sci. 2015; 30(4): 468-484.
- 17Fréchet M. Les éléments aléatoires de nature quelconque dans un espace distancié. Ann Inst H Poincaré. 1948; 10: 215-310.
- 18Lawton WH, Sylvestre EA, Maggio MS. Self modeling nonlinear regression. Technometrics. 1972; 14(3): 513-532.
- 19Kneip A, Gasser T. Statistical tools to analyze data representing a sample of curves. Ann Stat. 1992; 20(3): 1266-1305.
- 20Kneip A, Engel J. Model estimation in nonlinear regression under shape invariance. Ann Stat. 1995; 23(2): 551-570.
- 21Parati G, Rizzoni D, Omboni S, et al. The analysis of blood pressure and heart rate variability: methodological aspects and interpretation of results. High Blood Press. 1995; 4: 186-203.
- 22Winnicki M, Canali C, Mormino P, et al. Ambulatory blood pressure monitoring editing criteria: Is standardization needed? Am J Hypertens. 1997; 10(4): 419-427.
- 23Bigot J. Fréchet means of curves for signal averaging and application to ECG data analysis. Ann Appl Stat. 2013; 7(4): 2384-2401.
- 24Scholkopft B, Mullert K-R. Fisher discriminant analysis with kernels. In: Proceedings of the 1999 IEEE Signal Processing Society Workshop Neural Networks for Signal Processing IX. New York; 1999; Madison, WI, USA. 23-25.
