ABSTRACT

In radiation therapy, the majority ofMost radiation therapy researches have has focused on reducing unnecessary radiationdosage into normal tissues and critical organs around the target tumor volume. Proton beam therapy is considered asto be one of the most promising radiation therapy methods, withgiven its physical characteristics in the dose distribution, delivering most of the dose just before protons come to rest, at the so-named called Bragg peak;. thatThis is to say that, proton therapy allows for a very high radiation dose to the tumor volume, effectively sparing adjacent critical organs. To maximize these benefits and minimize the risk of misapplication of the Bragg peak, which is to say, to guarantee both successful treatment and patient safety, it is essential to accurately determine the distal dose edge and effectively monitoring the in vivo dose distribution. and precisely determining the distal dose edge is very important for the safety of patient and successful treatment. 

The present dissertation suggestsrecommends the prompt gamma measurement method tofor monitoring of the in vivo proton dose distributions during the treatment,. and the present study also verifiesAs a key part of the process of establishing the utility of that method, the verification of the clear relationship between the prompt gamma distribution and the proton dose distribution, was accomplished withby means of Monte Carlo simulations and experimental measurements.

First, the physical properties of prompt gammas were investigated basedon the basis of on cross-section data and Monte Carlo simulations. Prompt gammas are mainly generated mainly from the proton-induced nuclear interactions, and then emitted isotropically in less than 10-9 sec withat the energies of up to 10 MeV. in less than 10-9 sec. The sSimulation results offor the prompt gamma yield for the major elements of the human body showed that within the optimal energy range of 4-10 MeV, the highest number of prompt gammas areis generated from oxygen, whilewhereas forover the entire energy range, it is from calcium. generates the more prompt gammas than other elements.

Secondly, to seedetermine[JH1]  the relationship between the proton dose distribution and the prompt gamma distribution, the present study developed a proof-of-principle measurement system (the PGS system) employing thea scanning process. It isFirst-time experimentally  verifiedverification for the first time thatwas achieved,  not only of the fact that the prompt gammas can be measured during the treatment, but also that itstheir distribution has a clear relationship with proton dose distribution for the therapeutic energy range of proton beams.

Thirdly, for clinical application, a small array-type prompt gamma measurement system, which could be for used without the problematic scanning process, iswas designed, and its optimal dimensions tofor effectively reductione of background gammas arewere determined (by Monte Carlo simulations);: that is, the 3 mm scintillation thickness; of 3 mm, the 2 mm slit width; of 2 mm, the 2 mm collimation-plate thickness; of 2 mm, and the150 mm  collimation-plate length. of 150 mm are determined to be the optimal dimensions of the measurement system. To speed upaccelerate the simulations, the present study employeds a parameterized source term whichthat improveds the calculation speed up toby a factor of 300. times.

Finally, the performance of anthe array-type measurement system for the clinical application arewas examinedevaluated with thea test measurement system which is composedcomprising of a multi-slit collimation system, a CsI(Tl) scintillation detector, and a precise motion system. A methodology to determine the distal dose edge from the prompt gamma distribution is proposed in the present study. Additionally, the phantom effect on the prompt gamma distribution and the analysis of background gammas arewas studied by Monte Carlo simulations. A methodology for effective determination of the distal dose edge from the prompt gamma distribution is here proposed.

 

 


 

6.1 Conclusions

In this dissertation, thean in vivo dose verification method based on a prompt gamma measurement during proton therapy was suggestedrecommended, and its feasibility was verified with Monte Carlo simulations and experimental measurements.

First, the physical characteristics of the prompt gammas generated by the proton-induced nuclear interactions were investigated withusing cross-section data and Monte Carlo simulations. The present results show that the majority of prompt gammas are generated near the Bragg peak withat 4.44 MeV through the inelastic interactions of 16O(p,x)12C, 12C(p,x)11B, and 12C(p,x)12C, and thenafter which they it isare emitted isotropically in less than a nanosecond. The simulation study foron the prompt gamma yield with the major elements (carbon, oxygen, and calcium) and organs of athe human body showeds that calcium emits more than two times the number of prompt gammas more than two times thanthat carbon or oxygen do, when considering the entire energy range of 0-10 MeV prompt gammas energy range per one incident proton particle; however, ifwhen we considered only the optimal energy window of 4-10 MeV, the prompt gamma yield of oxygen was higher than that of calcium or carbon.

To experimentally measure the prompt gamma distribution and to seedetermine its relationship with the proton dose distribution, a proof-of-principle measurement system (the PGS system) was developed based on a scanning process. The PGS system was designed with the emphasies on the suppression of fast neutrons with three layers (the paraffin, B4C, and lead layers) and the selectively measurement of the gammas passing through only[JH2]  the collimation holeslit[JH3] . For the therapeutic proton beams, it iswas experimentally proved that the location of the distal dose edge in the water phantom couldcan be determined by measuring the prompt gamma distribution with the PGS system outside the water phantom with the PGS system.

For clinical purposes, a small array-type prompt gamma measurement system, which that could be used without the problematic scanning process, was designed, and its optimal dimensions tofor effectively reductione of the background gammas were determined withby the Monte Carlo method; to be that is, the3 mm scintillation thickness, of 3 mm, 2 mm the slit width, of 2 mm, the2 mm collimation-plate thickness, of 2 mm, and the150 mm collimation-plate length. of 150 mm were determined to be the optimal dimensions of the measurement system. For theTo reducetion of the computational burden in the optimization process, a parameterized source term offor the secondary particles[JH4]  was employed, andwhich reduced the computation time was reduced by a factor of up to 300 for each simulation case. The simulation results also confirmed that anthe array-type prompt gamma measurement system, in its optimal dimensions, wcould accurately determinelocate the location of the distal dose edge, (e.g., within a few millimeters of error) for the entire range of therapeutic proton beam energies.

Finally, the performance of anthe array-type measurement system was experimentally examined with the test measurement system withand the therapeutic proton beams. ThroughBy means of thea methodology based on the sigmoidal curve fitting, it was proved that with therapeutic proton beams within the 80 – 220 MeV energy range, the distal dose edge could be quantitatively determined within about 4 mm. with therapeutic proton beams of 80 – 220 MeV energy range. It is aAlso, it was revealed that the higher-energy of proton beams show the more neutron capture gammas and the moregreater beam dispersion, which are considered asto be the main problems in the prompt gamma measurement. The change[JH5]  of the prompt gamma distribution due to the phantom variation was studied by the Monte Carlo simulations. OurThe results showed that the phantom size, shape, and location do not significantly affect on the prompt gamma distribution. Although the phantom material variations showed some alteration in the prompt gamma distribution, the location of the distal dose edge did not change.


6.2 Future Work

Even though the results of the present study suggestedsuggested a methodology for the measurement of the prompt gamma distributions based onusing a small array-type measurement system, further studies are necessary forinvolving clinical use of the methodology in proton therapy are necessary. First, it is required that a full-scale array-type measurement system be constructed and evaluated with many different types of therapeutic proton beams. For more accurate prediction of the distal dose edge, the level of the background gammas should be as low as possible. It is expected that the full-scale array-type measurement system will yield better results, along with a lower level of the background gammas. Note that the current single-detector system, usesusing only one detector in a collimation slit, and therefore allows neutrons and photons to enter the measurement system through the other neighboring collimation slits without any interruption. In the full-scale measurement system, the neighboring collimation slits will be blocked with CsI(Tl) detectors, which will not only will provide some degree of the shielding effect from the incoming neutrons and photons, but also will enable the employment of the anti-coincidence counting. The anti-coincidence counting technique can be used to remove the scattering[JH6]  gammas coming from[JH7]  the neighboring scintillators. The prompt gammas also can also be exclusively measured exclusively by employingusing time gating, which is thea process of the selective measurement considering the time difference between the generation of prompt gammas (within 1 nanosecond) and the generation of the background capture gammas (within hundreds of microseconds).

Although ourthe test measurement system showed that the distal dose edge of the pencil beam (a single proton spot) could be predicted from a measured (longitudinal) distribution of prompt gammas, in order to achieve clinical applicability, it is also required that the two-dimensional (2D) distribution (not only the longitudinal but also the lateral distribution) of the proton dose covering the arbitrary shape of the tumor volume be determined. In the development of the 2D prompt gamma measurement system for the 2D proton dose distribution, it may be considered necessary that (1) the dimensions of the detector sensors arebe properly determined for both the effective measurement and the spatial resolution, (2) the energy calibration isbe performed automatically performed for a multi-channel measurement system of sufficient precision[JH8] , which could bewould[JH9]  consisted of more than 64 channels, with a sufficient precision, and (3) the advanced radiation detection techniques arebe employed to reduce the background gammas;  (for example, the particle tracking technology maymight not only reduce the gammas from unwanted directions, but could also increase the detection efficiency). In fact, physicians’ real-time visualization of proton dose distributions delivered to patients it maycould be an epoch-making discovery. that the proton dose distribution delivered to the patient is visualized in real-time for the medical staff and the patient; mMoreover, the combination of this real-time dose confirmation technology with the cutting-edge technology such as the Image Guided[JH10]  Radiation Therapy (IGRT) and the Adaptive Radiation Therapy (ART) technologies would provide greatsignificant additional benefits forto the patients treated with proton beams.

 


 [JH1]OR:

(1) confirm

(2) clarify

 [JH2]implicit

 [JH3](?) OR: slits

 [JH4]OR:

“secondary particle source term”

 [JH5]OR: variation

 [JH6]OR: scattered

 [JH7]OR:

“the gammas scattering [OR: scattered] from”

 [JH8]*OR:

“energy calibration be performed automatically and with sufficient precision for a multi-channel measurement system…”

 [JH9]OR:

“could possibly”

 [JH10]OR (if applicable): Image-Guided [hyphenated]