The nominal values of electron density listed in the phantom data sheets were 0.004, 2.851, 3.160, 3.343, 3.335, 3.833, 4.557, 6.243 (10^23 e/cm^3) for Air, PMP, LDPE, Water, Polystyrene, Acrylic, Delrin, and Teflon respectively. Tables 1 and 2 show the absolute and relative errors between the nominal and calculated values. The absolute errors for PMP, LDPE, Water, Polystyrene, Acrylic, and Delrin, excluding Air and Teflon, were within 0.066 (10^23 e/cm^3), and the relative errors were within 1.98%. The electron density values were calculated with high accuracy. From these results, the accuracy of the electron density value estimation is expected to be high for materials with an electron density value from about 3 to 5 (10^23 e/cm^3) and a CT value from -200 to 400 HU. On the other hand, the relative error of Air was significantly larger (-2000 to -3000%) than that of the other materials, while the absolute error was at most 0.123 (10^23 e/cm^3). This result was considered to be due to the low electron density value of 0.004 (10^23 e/cm^3) of Air. The relative error of Teflon was 3.00%, which was larger than that of other materials. However, the CT value of Teflon is high (1000 HU), so the accuracy of electron density value estimation may be inferior for materials with high CT values.


In this study, the electron density values at tube currents of 50, 100, 200, and 250 mA were obtained, and the dose dependence of electron density value was examined. Fig 3 shows the nominal values and the calculated values at each tube current for each material. The standard deviations for different tube currents were 0.016, 0.016, 0.003, 0.006. 0.023, 0.010, 0.007, 0.012 (10^23 e/cm^3) for Air, PMP, LDPE, Water, Polystyrene, Acrylic, Delrin, and Teflon, respectively, and the variation of electron density value due to the difference in tube current is small for all materials. Fig 4 shows scatter plots of the electron density value with the nominal value on the x-axis and the calculated value on the y-axis. The calculated values have a high correlation with the nominal values for all tube current conditions (r > 0.99), proving that the electron density values were calculated with high accuracy independent of dose.
On the other hand, the calculated values in this study tend to be about 0.4 to 3.4% larger than the nominal values. Future studies are required to examine estimation with higher accuracy and measurement using materials that constitute the human body.

