Several phantom models are available for representing the human body in Monte Carlo calculations (e.g., Kramer et al 1982, Cristy and Eckerman 1987, ICRU 1992a, Lee et al 2006a, Schlattl et al 2007). These include voxel-phantoms, which are based on CT and MR images of actual human beings, and computational models where body contours and organs are defined by mathematical expressions.

The phantoms used in PCXMC are computational hermaphrodite phantoms representing human beings of various ages: new-born, 1, 5, 10, 15-year-old and adult patients. The phantoms include expressions describing various organs and body parts. These phantoms have been specified by Cristy and Eckerman (1987), but a few modifications, explained below, have been made in PCXMC. The principal body dimensions of these phantoms are given in Table 3, and the composition of their tissues is shown in Table 4. The phantom models used in earlier versions of PCXMC were somewhat different: they were modelled according to Cristy (1980) with small modifications (Tapiovaara et al. 1997).

Table 3: Principal dimensions of the mathematical phantoms as modified in PCXMC. In the calculation the user can specify whether the arms of the phantom are included at the sides of the trunk or whether they are removed (which may simulate the real situation better, e.g., for lateral projections). Trunk width is given for both of these conditions. Phantom height and weight have changed from the earlier PCXMC versions.


 

Weight
(kg)

Total height
(cm)

Trunk height
(cm)

Trunk thickness
(cm)

Trunk width*
(cm)

Trunk width**
(cm)

Leg length
(cm)

Newborn

3.40

50.9

21.6

9.8

10.94

12.7

16.8

1 year old

9.20

74.4

30.7

13.0

15.12

17.6

26.5

5 year old

19.0

109.1

40.8

15.0

19.64

22.9

48.0

10 year old

32.4

139.8

50.8

16.8

23.84

27.8

66.0

15 year old

56.3

168.1

63.1

19.6

29.66

34.5

78.0

Adult

73.2

178.6

70.0

20.0

34.40

40.0

80.0

* excluding arms
** including arms

Table 4. Elemental composition of the phantom tissues as used in PCXMC (% by weight). The number of elements has been reduced from that in Cristy and Eckerman ( 1987) by grouping the elements Na, Mg, P, S, and Cl together and treating them as phosphorus, and grouping all elements of atomic numbers from that of K or higher together and treating them as calcium. The density and composition have changed from the earlier PCXMC versions.


 

Density
(g/cm3)

H
(%)

C
(%)

N
(%)

O
(%)

P
(%)

Ca
(%)

Skeleton (except newborn)

1.40

7.337

25.475

3.057

47.893

5.867

10.362

 Newborn skeleton

1.22

7.995

9.708

2.712

66.811

4.623

8.151

Lung tissue

0.296


 

10.134

10.238

2.866

75.752

0.770

0.240

Other soft tissues (except newborn)

1.04

10.454

22.663

2.490

63.525

0.626

0.242

Other soft tissues (newborn)

1.04

10.625

14.964

1.681

71.830

0.592

0.308

 

The phantom models of Cristy and Eckerman (1987) were intended to be used for dosimetry of internal photon sources.  After their publication, the head and neck region and the upper part of the spine of the phantoms have been modified by Eckerman and Ryman (1993) in order to have the phantom models better suited for external irradiation calculations. These modifications are explained in more detail in Eckerman, Cristy and Ryman (1996). Eckerman and Ryman (1993) also added the oesophagus among the organs in the phantoms. These modifications have also been included in PCXMC 2.0, although with small further modifications:

  • The back of the head has been modelled as a circular cone, rather than as an elliptic cylinder as in Eckerman, Cristy and Ryman (1996). For an example of the old and new head and neck area, see figure 1. The head model of PCXMC resembles, but is not equal to the MIRD head model (Bouchet et al 1999). 

  • The lateral width of the facial skeleton has been reduced from the value given in Cristy and Eckerman (1987) in order to make room for the parotid glands.

  • The apparent error in the vertical location of the facial skeleton in the data of Eckerman, Cristy and Ryman (1996) has been corrected, and the facial skeleton is located in a lower position than it would be according to their data.

  • The apparent error in the position of the thyroid in the data of Eckerman, Cristy and Ryman (1996) has been corrected. The data used in PCXMC for the adult phantom corresponds to the data given in Eckerman and Ryman (1993).

  • Salivary glands (parotid, sublingual and submandibular glands) have been modelled in the phantoms. The size and location of the glands were derived from data in ICRP Publication 89 (ICRP 2002) and Möller and Reif (1994a).

  • Extrathoracic airways (pharynx, larynx, part of trachea, paranasal sinuses) have been modelled in the phantoms. Guidance for the modelling was obtained from Möller and Reif (1994a). Mouth has not been considered as a part of extrathoracic airways, although it is among the extrathoracic airway tissues mentioned in ICRP Publication 89 (ICRP 2002). Mouth mucosa is treated separately in PCXMC.

  • Mouth mucosa has been modelled in the phantoms. Part of the tissue is located between the skin and the facial skeleton, and another part behind the facial skeleton.

  • The prostate has been modelled in the phantoms. Guidance for the modelling was obtained from ICRP Publication 89 (ICRP 2002) and Möller and Reif (1994b).

  • The arms of the phantoms can be removed in order to enable more realistic calculations for lateral x-ray projections. When the user chooses this option, it is realised by cutting the trunk region with two planes, which are parallel to the y-z plane, and located at the maximum dimension of the rib cage in the direction of the x-axis, but increased by the thickness of the skin. (The origin of the coordinate system is located at the middle of the base of the trunk of each phantom; the z-axis points upwards, the x-axis to the left-hand side of the phantom, and the phantom looks in the negative y-direction.)

  • The height and mass of the phantoms can be varied and matched to the data of individual patients. This feature is explained in more detail below.

  • Lymph nodes have not been modelled in the phantoms. Instead, the dose in the lymph nodes is estimated from the doses in surrogate organs as

Dlymph nodes = 0.25.Dsmall intestines + 0.15.Dpancreas        + 0.13.Dextrathoracic airways +

                     0.10.Dgall bladder     + 0.08.Dsalivary glands + 0.07.Dlungs +                           (2)

                     0.05.Dthyroid              + 0.05.Dtotal body        + 0.04.Doesophagus +

                     0.04.Dheart                 + 0.03.Dstomach          + 0.01.Dtestes .

These surrogate organs and the weights in the averaging have been chosen with the guidance of the data given in Qatarneh et al. (2006), Möller and Reif (1994a and 1994b) and ICRP Publication 89 (ICRP 2002).

a)

b)

Figure 1. An example of the head and neck area of the adult phantom, displayed with the “radiograph”-image in (a) PCXMC 2.0 (new model), (b) PCXMC 1.5 (modelled according to Cristy 1980). The head and neck model in the new version (2.0) is notably more appropriate for external irradiation dose calculations. The model of Eckerman, Cristy and Ryman (1993) would be in between these two models and would have excess soft tissue material at the back of the head: their head model is an elliptical cylinder topped by half an ellipsoid. The skeleton is shown in white, the thyroid in pink, the airways in blue, and the oesophagus in yellow colour.

Figure 2: Anterior views of the basic phantom models in PCXMC, scaled to have identical heights. (a) Adult phantom 178.6 cm/73.2 kg, (b) 15-year old phantom 168.1 cm/56.3 kg, (c) 10-year old phantom 139.8 cm/32.4 kg, (d) 5-year old phantom 109.1 cm/19.0 kg, (e) 1-year old  phantom 74.4 cm/9.2 kg (f) new-born phantom 50.9 cm/3.4 kg.

Figure 2 illustrates the exterior shapes of the basic phantom models. PCXMC allows further modification of these basic phantoms by letting the user change the mass (M) or height (h) of any of them. Using these target body size values, the program calculates scaling factors

        (2)

and

       (3)

where sz is the scaling factor in the direction of the z-axis (phantom height), sxy the scaling factor in the directions of the x- and y-axes (phantom width and thickness, respectively), and h0 and M0 are the height and weight of the unscaled phantom (Table 3). All dimensions of the phantoms are then multiplied by these scaling factors, and the organ masses are changed accordingly. This operation allows the shape of the phantoms to be modified to resemble that of the actual patient more closely. It should be noted, however, that all measures in a given direction (i.e., vertical or horizontal) are being multiplied by the same scaling factor, and phantom shape variability due to, e.g., variability in the amount of fat tissue cannot be simulated by this method. An example of the height and weight transformation of the phantom is given in Figure 3. This transformation does not have any effect on the coordinate system used for entering the x-ray beam geometry: the origin remains at the centre of the base of the trunk. The location of the organs will change, and the input coordinates of the x-ray beam have to be changed accordingly in order to simulate the irradiation of the same body part. The coordinates corresponding to the transformed phantom, (x, y, z), are obtained from the coordinates of the same point corresponding to the basic phantom, (xo, yo, zo), by

x = sxy xo

y = sxy yo                 (4)

z = sz zo

Figure 3. An example of the effect of the patient size transformation. The phantoms shown represent 5-year-old patients of height 109.1 cm with various mass: (a) 12.45 kg, (b) 19.0 kg, (c) 29.0 kg. The figure shows the body outline in the AP direction and the location of internal organs and the skeleton.

Jaa tämä sivu