J. A. Garcia M. Rubifio J. Romero E. Hita zyxwv YXWV DCBA Departamento de Optica Universidad de Granada 18071 Granada Spain Measuring the of Human-Teeth Whiteness zy hg SR This work attempts to determine the best index for measur- ing the whiteness zyxwvutsr GFEDCBA of human teeth. For this the classifica- tions obtained from various whiteness indices have been compared, using Spearman’s correlation coefficient, with the responses of 20 observers judging a group of 20 sam- ples oj‘porcelain teeth. The index which gives the best results is that proposed by the CIE, WIC, although the zyxwvu KJIHG Z% and the ClELAB color difference A E are also adequate for measuring the whiteness of human teeth. zyxwvut DCBA 0 1993 John Wiley zyxwvutsrqp & Sans, Inc. INTRODUCTION There are many industries in which the whiteness index is a useful measure of the manufactured products. De- pending on the kind of product one of several whiteness indices or other related indices has been proposed through history. For example, the paper industry is accustomed to using Z%, a whiteness index which takes into account the tendency for paper to yellow. If the task is to evaluate the whiteness of any other product, we must begin by selecting the most adequate whiteness index. Whitening teeth for esthetic purposes currently appears to be fashionable, causing many patients to seek profes- sional help in modifying tooth color. In recent years vari- ous whitening systems have appeared on the market, the results of which are almost always evaluated subjectively. The patient, for example, might compare photographs taken before and after the treatment, but very seldom do these evaluations involve colorimetric parameters. It is necessary that the industries which manufacture “whiten- ing systems” and dental prosthesis use an index which adequately evaluates whiteness and its variation in hu- man teeth. 0 1993 by John Wiley & Sons, Inc Volume 18, Number 5, October 1993 For an objective evaluation of their whitening treat- ment, Rosenstiel e f al.’ used the CIELAB color difference AE between sample colors before and after treatmen Nevertheless, this parameter is not necessarily definitive since a favorable alternative is to use the color differenc between the dental sample to be studied and a surface used as the whiteness model. In addition, we might propo evaluating this dental whitening by means of one of many whiteness indices which are currently used in industry and which have emerged to solve specific problems. In view of the above, our aim has been to determi which method best enables us to evaluate the whitene of teeth. For this, we have analyzed different parameters which classify dental whiteness, comparing the results o these with the classification of different observers. T method used for selecting the index in this case would be applicable if it deals with other kinds of samples. In the field of dental colorimetry there are few refer ences, (with exceptions, e.g. Nakaura et al.,*) which eval- uate the color coordinates of a series of dental samp in order to establish a color space model on tooth-crown color. Their results indicate the zone in Munsell atlas where tooth colors would be. WHITENESS INDICES Judging the whiteness of a given surface depends, among other things, upon the illumination under which the su face is observed. Therefore the best way to classify th degree of whiteness is in relation to other white surface Our definition of whiteness is from Wyszeckiand Stiles3: “. . . it is the attribute of a reflecting surface under a given illumination, which determines its similari with respect to a preferred or model white.” Accordin to this, to evaluate the whiteness of teeth, we must begin by selecting the appropriate model white and the illumin CCC 0361 -231 7/93/050349-04 349