Objective. The purpose of this study was to describe a new data visualisation system by plotting the human face to observe the comprehensive effects of multivariate data. Methods. The Graphics Device Interface (GDI+) in the Visual Studio.NET development platform was used to write a program that enables facial image parameters to be recorded, such as cropping and rotation, and can generate a new facial image according to values from sets of normal data ( was still counted as 3). The measured clinical laboratory parameters related to health status were obtained from senile people, glaucoma patients, and fatty liver patients to illustrate the facial data visualisation system. Results. When the eyes, nose, and mouth were rotated around their own axes at the same angle, the deformation effects were similar. The deformation effects for any abnormality of the eyes, nose, or mouth should be slightly higher than those for simultaneous abnormalities. The facial changes in the populations with different health statuses were significant compared with a control population. Conclusions. The comprehensive effects of multivariate may not equal the sum of each variable. The 3 facial data visualisation system can effectively distinguish people with poor health status from healthy people. 1. Introduction Data visualisation is the study of visual representation of data to communicate information clearly and effectively through graphical means [1–3]. The medical sciences have a uniquely intertwined relationship with bioinformatics. The rapidly expanding field of biology creates enormous challenges to enable researchers to gain insights from large and highly complex data sets. Although researchers and practitioners often create patterns that can be visually identified, such as charts, graphs, and interactive displays, when solving a large range of problems, there are no definite accepted methods to identify these complex relationships [4–6]. Traditional data visualisation systems are mostly based on mathematical models, but complex bioinformatic correlations may not follow previously known statistical rules. Therefore, it is essential to explore methods for data visualisation that do not completely rely on mathematical models. We established the facial data visualisation system based on changes in human facial features. Certain specific bioinformatics rules for correlations may be elucidated with the use of the data visualisation system. Chernoff first developed the idea of using human facial characteristics as a means to visualise data [7, 8]. The idea behind using faces is
References
[1]
S. Durinck, J. Bullard, P. T. Spellman, and S. Dudoit, “GenomeGraphs: integrated genomic data visualization with R,” BMC Bioinformatics, vol. 10, article 2, 2009.
[2]
A. P. Francisco, C. Vaz, P. T. Monteiro, J. Melo-Cristino, M. Ramirez, and J. A. Carrio, “PHYLOViZ: phylogenetic inference and data visualization for sequence based typing methods,” BMC Bioinformatics, vol. 13, article 87, 2012.
[3]
F. Jourdan, L. Cottret, L. Huc et al., “Use of reconstituted metabolic networks to assist in metabolomic data visualization and mining,” Metabolomics, vol. 6, no. 2, pp. 312–321, 2010.
[4]
J. Kennedy and J. Roerdink, “Highlights of the 1st IEEE symposium on biological data visualization,” BMC Bioinformatics, vol. 13, Supplement 8, article S1, 2012.
[5]
C. W. Bartlett, S. Y. Cheong, L. Hou et al., “An eQTL biological data visualization challenge and approaches from the visualization community,” BMC Bioinformatics, vol. 13, supplement 8, article S8, 2012.
[6]
J. H. Ostroff and D. C. Trost, “MDV: a multivariate data visualization tool for clinical laboratory data and other time-varying continuous measurements,” AMIA Annual Symposium Proceedings, vol. 1068, 2005.
[7]
H. Chernoff, “The use of faces to represent points in K-dimensional space graphically,” Journal of the American Statistical Association, vol. 68, no. 342, pp. 361–368.
[8]
B. Flury and H. Riedwyl, “Graphical representation of multivariate data by means of asymmetrical faces,” Journal of the American Statistical Association, vol. 76, no. 376, pp. 757–765, 1981.
[9]
ó. Kristjánsdóttir, A. M. Unruh, L. McAlpine, and P. J. McGrath, “A systematic review of cross-cultural comparison studies of child, parent, and health professional outcomes associated with pediatric medical procedures,” Journal of Pain, vol. 13, no. 3, pp. 207–219, 2012.
[10]
A. Chalmers, S. Harrison, K. Mollison, N. Molloy, and K. Gray, “Establishing sensory-based approaches in mental health inpatient care: a multidisciplinary approach,” Australasian Psychiatry, vol. 20, no. 1, pp. 35–39, 2012.
[11]
R. Lopez and B. Goldoftas, “The urban elderly in the United States: health status and the environment,” Reviews on Environmental Health, vol. 24, no. 1, pp. 47–57, 2009.
[12]
H. Liu, G. Wang, G. Luan, and Q. Liu, “Effects of sleep and sleep deprivation on blood cell count and hemostasis parameters in healthy humans,” Journal of Thrombosis and Thrombolysis, vol. 28, no. 1, pp. 46–49, 2009.
[13]
W.-F. Teng, W.-M. Sun, L.-F. Shi, D.-D. Hou, and H. Liu, “Effects of restraint stress on iron, zinc, calcium, and magnesium whole blood levels in mice,” Biological Trace Element Research, vol. 121, no. 3, pp. 243–248, 2008.
[14]
H. Liu, Y. Wang, X. Qi, and H. Yuan, “Serum glucose- and C-reactive protein-based assessment of stress status in a healthy population,” Clinical Laboratory, vol. 56, no. 5-6, pp. 227–230, 2010.
[15]
W. Hongwei, Z. Xinyu, L. Guihong, L. Xiliang, and L. Hui, “Nonspecific biochemical changes under different health statuses and a quantitative model based on biological markers to evaluate systemic function in humans,” Clinical Laboratory, vol. 56, no. 5-6, pp. 223–225, 2010.
[16]
L. Hui, L. Shijun, Z. Xinyu, W. Yuai, and X. Xiaoting, “Objective assessment of stress levels and health status using routinely measured clinical laboratory parameters as biomarkers,” Biomarkers, vol. 1, no. 6, pp. 525–529, 2011.
[17]
G. Lippi, G. Targher, M. Franchini, and G. C. Guidi, “Biochemical correlates of lipoprotein(a) in a general adult population. Possible implications for cardiovascular risk assessment,” Journal of Thrombosis and Thrombolysis, vol. 27, no. 1, pp. 44–47, 2009.
