Improving the robustness of the beam-scanning delivery in - - PowerPoint PPT Presentation

Improving the robustness of the beam-scanning delivery in proton-therapy.

Paul Morel March, 25 2014 LIGM - Laboratoire d’Informatique Gaspard Monge

Advisors: Guillaume Blin, PhD (LIGM, Universit´ e Paris-Est Marne La Vall´ ee, France) St´ ephane Vialette, PhD (LIGM, Universit´ e Paris-Est Marne La Vall´ ee, France) Xiaodong Wu, PhD (University of Iowa, USA)

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 1 / 38

Outline:

Proton-Therapy Proton-Therapy Simulator Motion Compensation

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 2 / 38

Proton-Therapy

Proton-Therapy

Cancer treatment relying on ionizing radiations aiming at killing cancerous cells using proton beams.

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 3 / 38

Proton-Therapy Why protons?

Why protons?

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 4 / 38

Proton-Therapy Why protons?

Why protons?

Figure 1: Comparison of spinal fields for medullblastoma: photons (upper panels) versus protons (lower panels) [Kirsch and Tarbell, 2004]

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 5 / 38

Proton-Therapy Why not only protons?

Why not only protons?

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 6 / 38

Proton-Therapy Protons therapy - an analogy

Protons therapy - an analogy

Water pressure = Proton energy ( depth of Bragg Peak) Water quantity = Dose = Number of protons Water drops are deposited on the way = Dose is cumulative along the way

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 7 / 38

Proton-Therapy Protons therapy - an analogy

Protons therapy - an analogy

Water pressure = Proton energy ( depth of Bragg Peak) Water quantity = Dose = Number of protons Water drops are deposited on the way = Dose is cumulative along the way

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 7 / 38

Proton-Therapy Protons therapy - an analogy

Protons therapy - an analogy

Water pressure = Proton energy ( depth of Bragg Peak) Water quantity = Dose = Number of protons Water drops are deposited on the way = Dose is cumulative along the way

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 7 / 38

Proton-Therapy Protons therapy - an analogy

Protons therapy - an analogy

Water pressure = Proton energy ( depth of Bragg Peak) Water quantity = Dose = Number of protons Water drops are deposited on the way = Dose is cumulative along the way

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 7 / 38

Proton-Therapy Protons therapy - an analogy

Protons therapy - an analogy

Water pressure = Proton energy ( depth of Bragg Peak) Water quantity = Dose = Number of protons Water drops are deposited on the way = Dose is cumulative along the way

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 7 / 38

Proton-Therapy Delivery Techniques

Main Principle: Energy Layers

Figure 2: Energy layers

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 8 / 38

Proton-Therapy Delivery Techniques

Pencil Beam Scanning

Discrete scanning: The beam is turned off between the spot positions.

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 9 / 38

Proton-Therapy Treatment Planning

Treatment Planning

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 10 / 38

Proton-Therapy Treatment Planning

Treatment Planning

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 11 / 38

Proton-Therapy Treatment Planning

Treatment Planning

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 12 / 38

Proton-Therapy Treatment Planning

Treatment Planning

weight ⇔ dose ⇔ amount of protons ⇔ duration

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 13 / 38

Proton-Therapy Motion

Sensitivity to motion

Inter-fraction motions: loss/gain of weight, tumor swelling/shrinkage, bladder, intestinal gas... Intra-fraction motions: breathing, heart beat ... ⇒ Interplay Effect

Figure 3: Results of irradiations without (left) and with (right) motion on a radiographic film.[Bert et al., 2012]

⇒ Overall treatment degraded

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 14 / 38

Proton-Therapy Motion

Sensitivity to motion

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 15 / 38

Proton-Therapy Motion

Motion mitigation I

During treatment planning: chose beam direction, use several beams, safety margins. Motion reduction:

Abdominal press Breath-Hold Anesthesia Breathing control techniques Beam-gating

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 16 / 38

Proton-Therapy Motion

Motion mitigation II

Rescanning (for interplay effect):

One energy slice: Iso-Layered Repainting Dose per spot visit is kept under an upper limit. It is characterized by tmax the time limit per spot per visit. Scaled Repainting The prescribed dose of every spots is divided by a constant N (repainting factor:number of repaintings). Applicable to the whole volume.

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 17 / 38

Proton-Therapy Motion

Motion mitigation III

A GPS for the body: Calypso (Varian)

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 18 / 38

Proton-Therapy Simulator

Proton-Therapy Simulator

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 19 / 38

Proton-Therapy Simulator Overview

Simulator overview

Main objective Quantify the impact of intra-fraction motions for given treatment plans. ⇒ Choice of the most robust plan. Implemented in Python with subroutines in C.

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 20 / 38

Proton-Therapy Simulator Patient Data

Patient data

Patient data: conversion CT # to mass densities and structure set information.

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 21 / 38

Proton-Therapy Simulator Physical Data

Physical Data

Depth dose curve generated from dose distributions in a water tank simulated in RayStation (RaySearch lab.) for energies ranging from 30MeV to 225MeV (step 5MeV). Missing energies: Approximation of the depth-dose curve from computed data.

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 22 / 38

Proton-Therapy Simulator Dose calculation

Dose calculation

Analytical model[Hong et al., 1996],[Szymanowski and Oelfke, 2002]: Main functions d(x, y, z) = Sm

w

ρm

w

C(z)O(x, y, z)

C(z) = DDw(rpl(z), E0) ∗ ssd0 + rpl(z) z 2

O(x, y, z) = 1 2π(σtot(z))2 ∗ exp

• −

x2 + y 2 2(σtot(z))2

• Auxiliary functions

σtot(z) =

• σ2

size + σpt(z)2

σpt(z) = y0(rpl(z))

y0(t) = y0(R) ∗

• 0.69 ∗

t R 2 + 0.33 t R

• y0(R) = 0.12085 × 10−4 ∗ R2 + 0.02275 ∗ R

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 23 / 38

Proton-Therapy Simulator Model evaluation

Model evaluation

Comparison to RayStation results: Single beamlets in water tank, energies from 30MeV to 225MeV: Dose profile and Bragg Peak: Comparison mean (mm) min(mm) max(mm)

• std. dev.(mm)
• RaySt. vs Simul.

1.1 2.73 1.15 Table 1: Absolute difference between the BP locations.

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 24 / 38

Proton-Therapy Simulator Model evaluation

Model evaluation

Comparison to RayStation results: Single beamlets in water tank, energies from 30MeV to 225MeV: Lateral profile at Bragg Peak: mean (mm) min(mm) max(mm)

• std. dev.(mm)

2.3 × 10−3 3.4 × 10−4 7.6 × 10−3 2 × 10−3 Table 2: Absolute difference between the std. dev. of the lateral profiles.

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 25 / 38

Proton-Therapy Simulator Model evaluation

Model evaluation

Treatment simulation:

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 26 / 38

Proton-Therapy Simulator Motion

Motion

We consider a patient moving during the treatment and being monitored: f (x, y, z, t) → (x′, y ′, z′)

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 27 / 38

Proton-Therapy Simulator Motion

Motion

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 28 / 38

Motion Compensation Compensation principle

Compensation principle

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 29 / 38

Motion Compensation Approaches Studied

Approaches Studied

Compensated Repainting Compensated unlimited rescanning with combining these methods:

Use margins for spot positions with or without map update. Figure 6: Map of original scanning positions (green) with a margin (red) for an energy layer.

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 30 / 38

Motion Compensation Experiments for 2D Motions

Experiments for 2D Motions

Experiment: Motion parameters:

X Y Z τ : Period(s) 4 3.5 A : Ampl (cm) 2 1.5 φ : Phase (rad)

(a) 2D Motion parameters

Delivery of 1 energy layer. Noise:

Amplitude: σx = 0.25cm, σy = 0.2cm Period: σx = 0.2s, σy = 0.1s

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 31 / 38

Motion Compensation Experiments for 2D Motions

Experiments for 2D Motions

Interest of the compensation method:

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 32 / 38

Motion Compensation Experiments for 2D Motions

Experiments for 2D Motions

Noise added to the measure provided by motion monitor: (σ = 0cm,σ = 0.4cm,σ = 1cm in both directions)

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 33 / 38

Conclusion

Conclusion

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 34 / 38

Conclusion

Conclusion

Current state: Simulate different treatment plans to render the effect of the motion. Dose distribution in the patient at each instant. Basic compensation technique: adapt the weight. Improvements for the future: Improve the compensation technique to be able to compensate at each unit

• f time.

Offer improvements for the treatment plan.

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 35 / 38

Acknowledgments

Acknowledgments

Members of the Dpt. of Radiation Oncology (University of Iowa, USA): Especially:

Dongxu Wang Ryan Flynn

Work partially supported by ANR project BIRDS JCJC SIMI 2-2010

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 36 / 38

Acknowledgments

Thank you!

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 37 / 38

Bibliography

Bibliography

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 38 / 38

Bibliography

Bert, C., Clasie, B. M., Craft, D., Engelsman, M., Flampouri, S., Flanz, J. B., Gottschalk, B., Ipe, N. E., Kooy, H. M., Li, Z., Lomax, A. J., Lu, H.-M., Paganetti, H., Palmans, H., Palta, J. R., Parodi, K., Schippers, M., Slopsema, R., Trofimov, A. V., Unkelbach, J., Van Luijk, P., and Yeung, D. K. (2012). Proton Therapy Physics. Taylor & Francis Group, LLC, series in edition. Hong, L., Goitein, M., Bucciolini, M., Comiskey, R., Gottschalk, B., Rosenthal, S., Serago, C., and Urie, M. (1996). A pencil beam algorithm for proton dose

• calculations. Physics in medicine and biology, 41(8):1305–30.

Kirsch, D. G. and Tarbell, N. J. (2004). Conformal radiation therapy for childhood CNS tumors. The oncologist, pages 442–450. Szymanowski, H. and Oelfke, U. (2002). Two-dimensional pencil beam scaling: an improved proton dose algorithm for heterogeneous media. Physics in Medicine and Biology, 47(18):3313.

Paul Morel Improving the robustness in Proton-Therapy March, 25 2014 38 / 38