QUALITY OF RADIOGRAPHIC IMAGES: LABORATORY EVALUATION OF INTRAORAL FILMS, FILTERS, COLLIMATORS, AND RADIATION EXPOSURE

QUALIDADE DA IMAGEM RADIOGRÁFICA: AVALIAÇÃO LABORATORIAL DE FILMES INTRA-ORAIS, FILTROS, COLIMADORES E EXPOSIÇÃO À RADIAÇÃO

QUALITY OF RADIOGRAPHIC IMAGES - LABORATORY EVALUATION OF INTRAORAL FILMS, FILTERS, COLLIMATORS, AND RADIATION EXPOSURE

QUALIDADE DA IMAGEM RADIOGRÁFICA - AVALIAÇÃO LABORATORIAL DE FILMES INTRA-ORAIS, FILTROS, COLIMADORES E EXPOSIÇÃO À RADIAÇÃO

José Roberto TAMBURUS^*

TAMBURUS, J. R. Quality of radiographic images: laboratory evaluation of intraoral films, filters, collimators, and radiation exposure. Rev Odontol Univ São Paulo, v. 11, n. 3, p. 161-167, jul./set. 1997.

In order to evaluate density, radiographic contrast and dose of radiation exposure, the author analyzed 80 radiographs containing 640 optical density data of the images of a penetrometer, exposed to the radiation beam with combinations between D and E periapical films, aluminum and copper/aluminum filters, and circular or rectangular collimators. The data obtained were analyzed by ANOVA and allowed the following conclusions: 1) aluminum filtration resulted in improved image contrast; 2) the use of group D film and an aluminum filter produced improved image contrast quality; 3) the rectangular collimator contributed to the production of improved contrast and to the reduction of radiation exposure, but did not affect density; 4) the combination of copper/aluminum filter, E group film and rectangular collimation significantly reduced radiation exposure.

UNITERMS: Image quality; Radiation filtration; X-ray film.

INTRODUCTION

Since X-rays were first discovered there has been concern about the control of their harmful effects. The evolution of studies and the progressive incorporation of new technical and scientific knowledge have led to the accumulation of information on how to control and/or reduce these effects without a loss in the quality of the radiographic image. Collimation and filtration of the radiation beam, control of the physical factors related to patient exposure, and the use of more sensitive radiographic films are the means used for this purpose.

With respect to film sensitivity, several studies comparing films of speed group D and E did not detect statistically significant differences in density or radiographic contrast, but noted that the use of E film permitted a reduction in the dose of exposure to radiation^6,20. Several investigators have studied the relationship between filtration of the radiation beam and quality of the radiographic image. REID¹⁸, RICHARDS¹⁹, and YALE; GOODMAN²⁹. studied the combination of aluminum and copper, both recommended by the International Commission on Radiological Protection for X-ray Filtration (TYNDALL²⁷, 1986). KOHN et al.¹² (1988) have compared the association between aluminum-yttrium and aluminum-copper filters; FARMAN et al.⁵ (1989) have studied the association aluminum-yttrium for X-ray filtration in intraoral techniques and have concluded that the addition of different filters such as copper might be used for extraoral techniques as well, although some loss of contrast can be expected using these combinations. Also in agreement with KOHN; et al.¹², the authors⁵ evaluated that copper should be preferable to any other as an additional filter. CORDT , ENGELKE² (1991) compared niobi and copper filters for intraoral radiographs. THORSEN et al.²⁴ (1990) carried out a comparison upon image quality, radiation dose reduction and tube load, following the use of either copper and niobi filters. Furthermore, the effects of circular and rectangular collimation of the radiation beam have also been studied in relation to radiation exposure and to quality of the radiographic image^{3,15,17,19,26}. The introduction in the marketplace of more sensitive films that were not available when REID¹⁸, RICHARDS¹⁹, and YALE; GOODMAN²⁹ conducted their studies and the fact that copper is an abundant low-cost element in nature, in contrast to rare metals, considered together with the possibility of reducing the radiation exposure were the factors that motivated the present investigation.

The objective of the present study was to investigate the density and contrast of the radiographic image obtained with speed group D and speed group E films, size 2, using two types of filters and two types of collimators for the radiation beam, as well as the doses of radiation exposure resulting from these combinations.

MATERIALS AND METHODS

The X-ray machine used was a Weber, model 11 R (Weber Dental Manufacturing Company, Canton 5, Ohio), calibrated to operate with 70 ± 2 kVp, 40 cm focus-film distance, and two filters, one of them consisting of 2 mm thick aluminum and the other consisting of 0.1 mm thick copper associated with 2 mm of aluminum. Kodak films of the D speed group with copper/aluminum filtration of the radiation beam were exposed for 2 seconds and E speed group films with the same filtration were exposed for 1.2 seconds. When the aluminum filter was used, D films were exposed for 1.5 seconds, and E films, for 0.7 seconds. In this laboratory study, an aluminum penetrometer was radiographed14 and the time of exposure used was preselected as a function of an optical density of approximately 1.0 obtained from the densitometry reading of the image of the fourth step (8 mm thickness) of the penetrometer. Effective kVp values, times of exposure and radiation doses were controlled with a NERO instrument (Non-invasive Evaluator of Radiation Outputs, model 6000B, Victoreen Inc.). In order to simulate the absorption of radiation by soft tissues, a 2 cm thick water layer was interposed between the radiation beam and the object/film assembly^17,19. Two millimeter thick lead collimators were positioned in front of the water layer. The circular one had an aperture of 50 mm in diameter and the rectangular one had an aperture of 32 x 42 mm (Figure 1).

Figure 1
- Diagrammatic view of the device used for exposition technique: 1 - Rx machine; 2 - focal spot; 3 - Pb collimator and filter; 4 - connectors; 5 - Pb diaphragm; 6 - water; 7 - room for penetrometer or mandibular segment; 8 - film; 9 - NERO sensor.

The films were processed manually under standardized conditions by the temperature-time method (20 ± 0.2ºC for 5 minutes) and dried in an oven circulating hot air. Eight sample groups were formed, corresponding to 8 combinations coded with the letters given in parenthesis: collimators - circular (A), rectangular (B); filters - 2 mm aluminum (X), 2 mm aluminum + 0.1 mm copper (Z); films - Ultraspeed (D), and Ektaspeed (E). Since each set consisted of 10 replications and since each replication contained the image of 8 steps of the aluminum penetrometer, 10 x 8 = 80 numerical data were obtained for each experimental group with respect to optical density. Since there were 8 experimental groups, there were 80 x 8 = 640 numerical data corresponding to the total sample studied. Densitometry readings were carried out using a digital Densitometer 0-424 (Victoreen, Inc.) with 1 mm diaphragm aperture, and the eight tables for the groups were constructed using log-transformed data. Sample data were not directly used for the evaluation of the characteristic contrast and optical density of each radiograph, but were submitted to mathematical transformation by the method of CAMPOS; TAMBURUS¹, which recommends hyperbolic transformation of the data to make them suitable for statistical treatment and analysis of the results. This transformation is made using the equation y = (a + bx)^-2 which corresponds to a second-order hyperbole. The final result shows that the experimental data, which originally followed a distribution according to a second-order hyperbolic curve, after the transformation above — in which the inverse of the square root of the original values (y' = y^-1/2) is taken as new experimental data —, tend to follow a linear distribution characterized by the following mathematical equation for the line: y = a + bx. In this equation, the b value that represents the slope tangent of the line expresses the radiographic contrast of the film and, the steeper the slope, the greater the b value and the greater the radiographic contrast. Thus, the distribution of b values can be used for the analysis of the experimental groups studied. The statistical tests to be used depend only on the type of distribution of b values, i.e., parametric tests for normal distribution and non-parametric tests for non-normal distribution. An additional transformation of the tangents can improve even more the probability of normal sample distribution or even lead to normalization of a non-normal distribution. This transformation initially involves the transformation of the original tangents into the corresponding circle arcs, and then, of these arcs into percent inclination of the curves. This was the procedure followed in the present study, which ultimately corresponds to the inverse of the transformation commonly employed in statistics, known as angular transformation, which leads to the transformation of originally percent data into arcsin, i.e., angles or arcs of a circle. This mathematical transformation is performed by first considering a w = arctang b value, with b being the same b value as for the regression line. From this we obtain:

w = arcsin (p/100)^1/2Þ sin (w) = (p/100)^1/2

sin² (w) = p/100 Þ p = 100 . sin² (w)

or p = 100. sin² (arctang b)

On the other hand, the a value of the equation for the line that indicates the point where the line cuts the ordinate axis (y), i.e., when x = 0, expresses the optical density of the film in the area located outside the surface of the film occupied by the penetrometer, i.e., at what we may call zero thickness, and therefore can be used to study and compare the optical density of the experimental groups.

RESULTS AND DISCUSSION

The linear regression resulting from the hyperbolic transformation of original values not only transformed the curve into a straight line, but also changed its direction from descending to ascending along the abscissa axis. In order to compare the optical densities resulting from the various combinations of film sensitivity/radiation filter/collimators, we used the same data pairs, but now, individually for each film. Thus, a linear regression test was performed for each radiograph taken, for a total of 80 tests (8 groups with 10 replications each), and 80 line equations were obtained for each film (y_i = a_i + b_i . x_i, with i varying from 1 to 80).

Radiographic optical density

The optical densities of the various experimental groups were compared by statistical analysis of the different a values on the 80 regression lines calculated.

To define the type of statistical analysis, a set of tests was performed to determine the normality of distribution of these values (H0 = 70.30). This set of tests demonstrated that optical density data had normal distribution, thus permitting the application of analysis of variance which showed statistical significance at the 1% level for the differences between films and between filters, and at the 5% for the differences between collimators. With respect to the interactions of the three variation factors, the test showed a significant difference at the 1% level for the film X collimator and film X filter interactions, and at the 5% level for the collimator X filter interactions. The results of analysis of variance and of comparison of the means demonstrated that: a) type E films (mean = 0.6657) presented a significantly lower optical density than type D films (mean = 0.6597); b) the copper/aluminum filter (mean = 0.6794) caused a statistically lower decrease in optical density than the aluminum filter (mean = 0.6461); and c) the difference between the 0.6661 mean value for the circular collimator and the 0.6603 mean value for the rectangular collimator was relevant at the 5% level of significance. It should be kept in mind that lower values represent higher densities due to the inversion of the data by hyperbolic transformation. These results agree with those reported by RICHARDS¹⁹, SPANGENBERG Jr.; POLL ²² and PREECE; JENSEN¹⁶, who, together with THUNTY; WEINBERG²⁵, defined the maximum and minimum limits of the optical density scales compatible with the range of variation that is useful in dental radiographic diagnosis. For these same authors, the copper/aluminum combination for filtration of the radiation beam resulted in optical densities outside these limits. SILHA²⁰ and THUNTY; WEINBERG²⁵ did not observe differences in optical densities between films, in contrast to the results reported here. With respect to the collimators, RICHARDS¹⁹ and HORNER; HIRSCHMANN⁹ observed that an increase in the collimation of the radiation beam reduced scattered radiation, thus contributing to the quality of the radiographic image by decreasing the fog.

Radiographic contrast

Eighty numerical data corresponding to the b values in the equations obtained for the radiographs studied here were used to evaluate the radiographic contrast determined by the combination of three factors. The fact that hyperbolic transformation converted the regression curves into straight lines, which only have one tangent represented by their own degree of inclination in relation to the horizontal plane, expressed by the b value in the mathematical equation of the line, allows a correct and simple comparison between lines. Because the original data did not present normal distribution, the transformations described earlier were performed (tangents of the slope angle to percent inclination of the regression lines). New tests performed after its treatment led to normal distribution (H0 = 29.09) which was sufficient to allow the application of parametric tests and analysis of variance. This analysis demonstrated statistically significant differences between films, between filters and between collimators, all at the 1% level of probability (p < 0.001) for the hypothesis of equality of the means. When the interactions between the three factors were tested, analysis of variance showed a significant difference at the 1% level of probability for the film X collimator interaction, a significant difference at the 5% level for the film X filter interaction (p < 0.05), and a nonsignificant difference for the filter X collimator interaction (p > 0.05). The results of analysis of variance and comparison of these means demonstrated that: a) type D films (mean = 0.1866) presented a significantly higher radiographic contrast than type E films (mean = 0.1411); b) the aluminum filter (mean = 0.1785) caused a significantly higher radiographic contrast in the films than did the copper/aluminum filter (mean = 0.1492); and c) the rectangular collimator (mean = 0.1751) caused a significantly higher contrast in the radiographic films than did the circular collimator (mean = 0.1526). It should be pointed out that hyperbolic transformation does not interfere with the interpretation of radiographic contrast since the latter depends only on the slope of the line and not on absolute values of optical density.

The film data obtained here agree with those reported by SILHA^20,21, PONCE et al.¹⁵ and THUNTY; WEINBERG²⁵ and disagree with those reported by FRYKHOLM⁶ and KAFFE et al.¹⁰. According to HORNER; HIRSCHMANN⁹, the filters should be compatible with intraoral film sensitivity, which is located in the photon range with energies of 35 to 55 keV. Aluminum is well suited for filtration of low-energy photons and for this reason several investigators recommend its combination with rare metals. SILHA²¹ stated that the addition of filters causes a loss of contrast because low-energy photons are removed from the beam; since these photons are less penetrating, their removal causes what he called a "hardening" of the beam, with a consequent loss of contrast. In contrast, HOLYOAK⁸ stated that inadequate filtration results in a "soft" radiation beam that only serves to increase the dose of radiation received by the patient without any improvement in the quality of the radiographic image. Some rare metals such as samarium require twice as much time of exposure compared to aluminum7 to produce the same optical density. YALE; GOODMAN²⁹ studied different thicknesses of the copper/aluminum combination for filtration, and concluded that it produces long contrast scales and substantially reduces radiation dose. RICHARDS¹⁹ used a copper thickness of 0.35 mm in combination with an aluminum thickness of 0.5 mm and observed an appreciable reduction in the level of contrast. He also observed that this combination prevented the full utilization of the variation in optical density that the film might potentially present, and that prolonged exposure times were required. REID¹⁸ also observed that filtration with copper/aluminum resulted in a higher contrast scale and this was not accepted by many professionals. Although in the present study more sensitive films were used which were not available at the time when the above investigators^12,18,19 carried out their research, and the copper thickness was reduced from 0.35 mm (RICHARDS¹⁹, 1958) to 0.1 mm with a significant reduction in exposure time, the present results show that loss of radiographic contrast continued to occur, in agreement with the results reported by the above authors^12,18,19. Thus, one should consider the observations of PONCE et al.¹⁵ who stated that the choice of radiographic contrast greatly depends on individual preference and that, if some loss in contrast scale is accepted, the copper/aluminum combination and group E film can be used to reduce the dose of exposure compared to aluminum filtration and group D film.

The opinions about the ideal and adequate aperture of the collimator for the radiation beam are divergent^3,8,16,17,19. The present study followed the guidelines of the COUNCIL ON DENTAL MATERIALS, INSTRUMENTS, AND EQUIPMENT^3,4 indicating that the irradiated area of the patient's face should not exceed the area of the periapical film. According to HORNER; HIRSCHMANN⁹ a reduction in collimator diameter contributes to the reduction of scattered radiation and consequently to the improvement of radiographic contrast. The present results demonstrated that the collimators had no significant effect on the optical density of the radiographs, but that the rectangular collimator contributed in a statistically significant manner to the improvement of radiographic contrast, in agreement with the above authors⁹.

Radiation dose

The data used in the evaluation of the exposure to radiation expressed in mR (mR = 10^-5 Gy) were determined in pilot experiments carried out in the laboratory and monitored with the aid of a NERO instrument, taking as a base the amount of radiation needed to produce an optical density of approximately 1.0 in the image of step 4 (8 mm thickness) of the penetrometer. The distribution of the data was found to be normal (H0 = 17.77). Analysis of variance showed that all factors involved and their interactions were significant at a level of less than 1%. With respect to collimation, the results demonstrated a significant reduction (p < 0.01) in the radiation dose needed to obtain intraoral radiographs when rectangular rather than circular collimation was used. This result agrees with the data reported by RICHARDS¹⁹, WHITE; ROSE²⁸, STENSTRÖM et al.²³ and KIRCOS; ANGIN¹¹. The results demonstrated that the copper/aluminum association used as filter significantly reduced the dose of radiation (p < 0.01) compared to aluminum filtration, in agreement with results reported by YALE; GOODMANN²⁹. With respect to film sensitivity, the present study showed a highly significant (p < 0.01) reduction in dose of exposure when E speed group films were used, in agreement with the data reported by FRYKHOLM⁶, PONCE et al.¹⁵, and KAFFE et al.¹⁰ and also with the recommendations of the COUNCIL ON DENTAL MATERIALS, INSTRUMENTS, AND EQUIPMENT^3,4. The results obtained here with respect to radiation dose may be summarized as follows, considering the means for the major factors as a term of comparison: a) E film <<< D film; b) copper/aluminum filter <<< aluminum filter; and c) rectangular collimator <<< circular collimator, with the "<<<" symbol (much lower than) indicating the high statistical significance of the results.

CONCLUSIONS

1. Filtration of the radiation beam with aluminum resulted in significantly higher image contrast compared to filtration with copper/aluminum.

2. With respect to optical density, the combination of group D film and aluminum filter presented a more adequate quality of the image.

3. Rectangular collimation produced a significantly more appropriate image contrast and reduction in dose of exposure to radiation; it had no significant effect on the image optical density.

4. When image contrast is an essential factor, the use of speed group D film, aluminum filtration and rectangular collimation is indicated.

5. Combinations of speed group E film, copper/aluminum filters and rectangular collimation significantly reduced the radiation exposure.

TAMBURUS, J. R. Qualidade da imagem radiográfica: avaliação laboratorial de filmes intra-orais, filtros, colimadores e exposição à radiação. Rev Odontol Univ São Paulo, v. 11, n. 3, p. 161-167, jul./set. 1997.

Com objetivo de avaliar a densidade, o contraste radiográfico e a dose de exposição à radiação, o autor analisou 80 radiografias contendo 640 dados de densidade óptica das imagens de um penetrômetro, exposto à radiação com combinações entre filmes periapicais dos grupos D e E de sensibilidade, com filtros de alumínio e cobre/alumínio e colimadores circular e retangular. Os dados obtidos foram analisados pela ANOVA e permitiram as seguintes conclusões: 1) a filtração com alumínio resultou em melhor qualidade de contraste das imagens; 2) o uso do filme do grupo D e do filtro de alumínio resultou em melhor qualidade do contraste; 3) a colimação retangular contribuiu para o melhor contraste e para a redução da exposição à radiação, porém, não alterou a densidade; 4) a combinação do filtro cobre/alumínio com filme do grupo E e colimação retangular reduziu significantemente a dose de exposição à radiação.

UNITERMOS: Qualidade da imagem; Filtração da radiação; Filme para raios X.

BIBLIOGRAFIC REFERENCES

Recebido para publicação em 26/03/96

Aceito para publicação em 07/04/97

• ¹
  CAMPOS, G. M.; TAMBURUS, J. R. A method to evaluate and compare roentgenograms. Braz Dent J, v. 2, p. 95-102, 1991.
• ²
  CORDT, I.; ENGELKE, W. Are additional filters made of niobium superior to the copper filter in dental radiology? Schweiz Monatsschr Zahnmed, v. 100, n. 10, p. 1160-1163, 1990.
• ³
  COUNCIL ON DENTAL MATERIALS, INSTRUMENTS, AND EQUIPMENT. Biological effects of radiation from dental radiography. J Am Dent Assoc, v. 105, n. 2, p. 275-278, 1982.
• ⁴
  COUNCIL ON DENTAL MATERIALS, INSTRUMENTS, AND EQUIPMENTS. Recommendations in radiographic practices, 1984. J Am Dent Assoc, v. 109, n. 5, p. 764-765, 1984.
• ⁵
  FARMAN, A. G. et al Evaluation of aluminum-yttrium filtration for intraoral radiography. Oral Surg Oral Med Oral Pathol, v. 67, n. 2, p. 224-226, 1989.
• ⁶
  FRYKHOLM, A. Kodak Ektaspeed: a new dental x-ray film. Dentomaxillofac Radiol, v. 12, n. 1, p. 47-49, 1983.
• ⁷
  GELSKEY, D. E.; BAKER, C. S. Energy-selective filtration on dental x-ray beams. Oral Surg Oral Med Oral Pathol, v. 52, n. 5, p. 565-567, 1981.
• ⁸
  HOLYOAK, B. Radiological protection in the dental profession. Br Dent J, v. 146, n. 6, p. 189-192, 1979.
• ⁹
  HORNER, K.; HIRSCHMANN, P. N. Dose reduction in dental radiography. J Dent, v. 18, n. 4, p. 171-184, 1990.
• ¹⁰
  KAFFE, I. et al Densitometric evaluation of intraoral x-ray films: Ektaspeed versus Ultraspeed. Oral Surg Oral Med Oral Pathol, v. 57, n. 3, p. 3342-3387, 1984.
• ¹¹
  KIRKOS, L. T.; ANGIN, L. L. Order of magnitude dose reduction in intraoral radiography. Oral Surg Oral Med Oral Pathol, v. 114, n. 3, p. 344-347, 1987.
• ¹²
  KOHN, M. L. et al Filters for radiation reduction: a comparison. Radiology, v. 167, p. 255-257, 1988.
• ¹³
  LUDLOW, J. B. Reducing the risk of dental x-ray exposure. J Mich Dent Assoc, v. 66, n. 7/8, p. 256-261, 1984.
• ¹⁴
  MANSON-HING, L. R.; BLOXOM, R. M. A stepwedge quality assurance test for machine and processor in dental radiography. J Am Dent Assoc, v. 110, n. 6, p. 910-913, 1985.
• ¹⁵
  PONCE, A. Z. et al Use of E-speed film with added filtration. Oral Surg Oral Med Oral Pathol, v. 61, n. 3, p. 297-299, 1986.
• ¹⁶
  PREECE, J. W.; JENSEN, C. W. Variations in film exposure, effective kVp, and HVL among thirty-five dental x-ray units. Oral Surg Oral Med Oral Pathol, v. 56, n. 6, p. 655-661, 1983.
• ¹⁷
  PRICE, C. The effects of kilovoltage, filtration and cone length on the attenuation of a dental x-ray beam by water. Dentomaxillofac Radiol, v. 10, n. 2, p. 59-63, 1981.
• ¹⁸
  REID, J. A. Improved exposure and processing techniques in dental radiology. J Canad Dent Assoc, v. 33, n. 9, p. 474-486, 1967.
• ¹⁹
  RICHARDS, A. G. Roentgen-ray doses in dental roentgenography. J Am Dent Assoc, v. 56, n. 3, p. 351-368, 1958.
• ²⁰
  SILHA, R. E. The new Kodak Ektaspeed dental x-ray film. Dental Radiogr Photogr, v. 54, n. 2, p. 32-35, 1981.
• ²¹
  SILHA, R. E. Methods for reducing patient exposure combined with Kodak Ektaspeed dental x-ray film. Dent Radiogr Photogr, v. 54, n. 4, p. 80-87, 1981.
• ²²
  SPANGENBERG Jr., H. D.; POOL, M. L. A 65 or a 90 kilovolt x-ray machine? Oral Surg Oral Med Oral Pathol, v. 13, n. 5, p. 552-565, 1960.
• ²³
  STENSTRÖM, B. et al Absorbed doses from intraoral radiography with special emphasis on collimator dimensions. Swed Dent J, v. 10, n. 1-2, p. 59-71, 1986.
• ²⁴
  THORSEN, F. et al The use of copper and niobium filters in diagnostic x-ray equipment: a comparison of dose reduction, tube loading and picture quality. Tidsskr Nor Laegiforen, v. 110, n. 22, p. 2878-2880, 1990.
• ²⁵
  THUNTHY, K. H.; WEINBERG, R. Sensitometric comparison of dental films of groups D and E. Oral Surg Oral Med Oral Pathol, v. 54, n. 2, p. 250-252, 1982.
• ²⁶
  TYNDALL, D. A. Rare earth filtration: spectral distribution, exposure reduction and image quality effects for panoramic radiography. Dentomaxillofac Radiol, v. 16, n. 1, p. 23-27, 1987.
• ²⁷
  TYNDALL, D. A. Spectroscopic analysis and dosimetry of diagnostic x-ray beams filtered by rare earth materials. Oral Surg Oral Med Oral Pathol, v. 62, n. 2, p. 205-211, 1986.
• ²⁸
  WHITE, S. C.; ROSE, T. C. Absorbed bone marrow dose in certain dental radiographic techniques. J Am Dent Assoc, v. 98, n. 4, p. 553-558, 1979.
• ²⁹
  YALE, S. H.; GOODMAN, L. S. Reduction of radiation output of the standard dental x-ray machine utilizing copper for external filtration. J Am Dent Assoc, v. 54, n. 3, p. 354-356, 1957.

*

Associate Professor, Radiology Course, School of Dentistry, University of São Paulo, Brazil.

Publication Dates

• Publication in this collection
  11 Mar 1999
• Date of issue
  July 1997

History

• Received
  26 Mar 1996
• Accepted
  07 Apr 1997