ESTRO 2024 - Abstract Book

S3941

Physics - Image acquisition and processing

ESTRO 2024

Figure 2 shows the SUVmax and MTV values for the device in the AP and CC motion, for the static and OncoFreeze AI cases, as well as the difference obtained between the two cases using Eq.1.

Figure 2. SUVmax and MTV for breathing phantom.

In patients, the MTV were classified according to: size, whether the lesion was larger than 1 cm3 or smaller than 1 cm3; and location, where lesions were chosen in the lungs and abdomen because of the influence the respiratory movement, and other lesions in areas without any motion, such as the pelvis.

Table 1 shows the SUVmax and MTV results for the 10 lesions with the corresponding differences between the two algorithms of reconstruction.

Static

OncoFreeze

ΔSUVmax (%)

ΔMTV (%)

SUVmax

MTV (cm3)

SUVmax

MTV (cm3)

29.84

0.33 0.52 0.31 2.27 9.08 1.44 3.41 3.75 2.80 0.42

41.61 10.33 10.40 18.51

0.20 0.39 0.18 2.31 8.63 1.13 2.23 2.69 2.60 0.40

39% 13% 21%

-39% -25% -42%

9.16 8.62

<1 cm3

Size

18.28 21.67

1% 2%

2%

>1 cm3

22.2

-5%

8.84

10.48 12.08

19% 16% 24%

-22% -35% -28%

With motion

10.41

7.94 9.06

9.86 9.60

Location

Without

6% 5%

-7% -5%

11.39

11.96

motion

Table 1. Patients with lesions of different size and location and their influence on SUVmax and MTV.

Regarding the results obtained from the phantom, motion in anterior-posterior direction is the most affected by respiratory movement, as shown by the major difference in the MTV. In terms of the patients, when the lesions are small (< 1 cm3) and move, the differences between OncoFreeze AI and static reconstructions become evident, in some case that differences are about 40% . On the other hand, when the lesions are larger o located in areas where there is no motion, there is a little difference between one method and the other.

Conclusion: