Conference PaperPDF Available

Proton Computed Tomography Imaging for Proton Radiation Therapy



CONCLUSION Proton CT has the potential to substantially improve the range accuracy of proton beams and to provide a low-dose imaging modality for daily image guidance. Careful evaluation of this novel technique is underway. BACKGROUND Proton CT is a novel tomographic imaging modality, which has become a realistic possibility with the increasing availability of rotating proton gantries and sufficiently high proton energies to penetrate the patient. Proton CT is hypothesized to improve the accuracy of proton treatment planning and to provide a low-dose method for daily image-guidance in proton therapy. In recent years, we have constructed a prototype proton CT scanner and installed it on a proton research beam line for further study . The scanner detects individual protons up to a maximum rate of about 105 protons/sec and records their entry and exit coordinates and directions with silicon strip detectors, as well as their residual energy by stopping them in an augmented 18-crystal CsI calorimeter. After calibrating the calorimeter response to water equivalent path length (WEPL), the scanner measures the integral of the relative stopping power (RSP) along the path of each proton The distribution of RSP values can then be reconstructed based on the WEPL values of many protons intersecting the object from multiple directions using advanced iterative algorithms based on projection onto hyperplanes. EVALUATION We have carefully calibrated the scanner with polystyrene plates of known thickness and RSP and obtained first images of a spherical quality assurance phantom (Lucy, Standard Imaging). We are now performing rigorous testing with CT QA phantoms (Catphan, The Phantom Laboratory) to compare characteristics including dose-fluence-noise relationship, sensitometry, spatial resolution, and low-contrast resolution with a state-of-the-art X-ray CT scanner. DISCUSSION While minimization of current proton range errors in the presence of tissue inhomogeneities is important for proton treatment planning, acceptable image quality at low doses is important for image guidance and target delineation, which will be evaluated within this project.
Proton Computed Tomography Imaging for Proton Radiation
Reinhard W. Schulte, MD, MS, Loma Linda University & Loma Linda University Medical Center
Disclosure Statement
Proton CT research is funded by Award
R01EB013118 from NIBIB to LLU, UCSC, CSUSB
and by Department of Defense Grant
W81XWH-10-1-017 to NIU
The speaker has no financial conflict
R Schulte, Proton Computed Tomography
Clinical & historical perspective: proton
therapy and proton imaging
Conceptual approach to proton CT
Technical realization of proton CT and first
R Schulte, Proton Computed Tomography
R Schulte, Proton Computed Tomography
Protons in Comparison
Proton energy deposition per track
length (proportional to dose)
increases as they slow down
Bragg peak dose at the end of the
proton range
Depth of peak (proton range)
adjustable by choosing right energy
Active or passive modulation
generates spread-out Bragg peak
Dose sparing up- and down-stream
from tumor
R Schulte, Proton Computed Tomography
Proton Therapy
Proton therapy was first used in a
hospital setting at LLUMC in 1990
More than 14,000 patients have
been treated at LLUMC since then
Due to the Bragg peak feature,
protons deliver less dose to normal
tissue than IMRT
As of June 2011, there were a total
of 37 proton therapy centers
worldwide, with many more under
R Schulte, Proton Computed Tomography
Proton Radiography (Andy Koehler, HCL)
A. Koehler was the first to point out
the potential value of proton
radiography and to perform
experiments with 160 MeV (Koehler,
Science 160, 303304, 1968)
The higher density resolution but
poorer spatial resolution was noted
by Koehler and later by Kramer et al.
(Radiology, 1980)
Interest in p-radiography as a QA
tool for proton therapy was revived
by U. Schneider at PSA during the
1990s R Schulte, Proton Computed Tomography
Andy Koehler, former director of the
Harvard Cyclotron
Proton Computed Tomography: Allan Cormack
(HCL, 1963)
Alan M. Cormack, physicist (1924-1998)
was the first to suggest reconstruction of
tomographic images based on X-ray
absorption and proton degradation (J.
Appl. Phys. 34, 2722, 1963)
It took less than 10 years before his idea
became reality when the first when
Geoffrey Hounsfield and colleagues
constructed the first X-ray CT scanner
Wide-spread of X-rays favored the rapid
development of X-ray CT
R Schulte, Proton Computed Tomography
Alan M. Cormack, 1924-1998 Physics
Nobel Laureate 1979, promoted the
idea of proton CT
Proton CT Collaboration
A small group of scientists first met at IEEE in Norfolk, VA in 2002 and BNL
in 2003 to outline the goal of building a clinical proton CT (pCT) scanner,
many others have contributed since then
Phase 0 (2003 2007): Conceptual design, Geant4 simulations, most likely
path concept, proof-of-principle experiments
Phase I (2008 now): First generation (preclinical) pCT scanner using
single particle tracking
Phase II (started in 2011): Second generation (clinical) pCT scanner
R Schulte, Proton Computed Tomography
Gensheimer F, et al., Int J Radiation Oncol Biol
Phys 2010 (50% idosdose, red = observed, blue =
R Schulte, Proton Computed
Motivation for pCT
Range Uncertainties in Proton Therapy
Differences in the interaction of x-rays
and protons with matter make proton
range calculations uncertain
Range uncertainties can range from mm
to cm
Materials of unknown stopping power
and CT artifacts create additional
Proton CT is a potential solution to
reduce this uncertainty from 3%-4% to ≤
1% of range
R Schulte, Proton Computed Tomography
Single-Proton pCT Concept
An energetic low intensity
cone beam of protons
traverses the patient
The position and direction
(entry & exit) and energy
loss of each proton is
Proton histories from
multiple projection angles
Minimal proton loss and
high detection efficiency
make this a low-dose
imaging modality
R Schulte, Proton Computed Tomography
Design of a Proton CT Scanner rotating with the
proton gantry (R Schulte et al. IEEE Trans. Nucl. Sci., 51(3), 866-872, 2004)
Low intensity
proton beam
Tracking of
pCT Imaging Process
The measurement is proportional to the outgoing energy of each
proton, and thus its energy loss in the phantom along its path
The energy loss can be converted to a line integral of proton stopping
power relative to water (RSP) along the proton path, RSP is (practically)
independent of proton energy
Energy straggling and detector noise limits RSP resolution
Multiple Coulomb scattering of protons limits spatial resolution
Most likely path estimation of protons within the object can improve
spatial resolution
Algebraic reconstruction algorithms can handle non-linear paths =>
iterative projection methods, e.g., ART, BIP
R Schulte, Proton Computed Tomography
R Schulte, Proton Computed
First Iterative Algorithm Applied: Algebraic
Reconstruction Technique (ART) (Tianfang Li &
Jerome Liang, SUNY, 2006)
ART was originally proposed
by Kaczmarz, 1937
Sequential orthogonal
projections onto
Works well for pCT but is
inherently slow
High frequency noise due to
reconstruction is apparent Li et al Med Phys 2006
R Schulte, Proton Computed
Further Development of pCT Reconstruction with
Geant4 Simulations (Scott Penfold, PhD Thesis)
Simulate pCT system and digital head phantom1
Test/compare/refine different reconstruction algorithms
1G. T. Herman, Image Reconstruction From Projections: The Fundamentals of Computerized
Tomography, Academic Press, New York (1980).
R Schulte, Proton Computed Tomography
Phase I pCT Scanner Design
(completed in 2010)
Horizontal beamline
Rotational stage for
object rotation
Upstream and
downstream tracker
Downstream energy
detector (calorimeter)
R Schulte, Proton Computed Tomography
Phase I pCT Scanner at LLUMC: Timeline
System component integration &
mounting (April 2010)
Testing with radioactive source and
cosmic rays (muons)
Installation on proton research beam
line & 1st test runs (May 2010)
Spill uniformity optimization (June
Scanner calibration (July 2010)
Phantom scans since Dec 2010
R Schulte, Proton Computed Tomography
Phase I pCT scanner with Rando head phantom
Proton CT Reconstruction
A series of pCT scans of
the Lucyradiosurgery
phantom was started in
Oct 2010
Phantom images were
reconstructed with
newly developed 3D
Image quality
progressively improved
R Schulte, Proton Computed Tomography
Phase I pCT Reconstruction of Lucy
R Schulte, Proton Computed Tomography
14 cm diameter polystyrene sphere with tissue equivalent inserts
64 slice Multi-Slice X-ray CT, 0.53 x 0.53 x
0.625 mm3
pCT Phase I, 0.63 x 0.63 x 2.5 mm3
Phase I pCT Reconstruction of the
CatphanUniformity Module
R Schulte, Proton Computed Tomography
A series of pCT scans of
standard CT performance
phantom modules is underway
The uniformity module tests
hypothesis that pCT RSP
reconstruction is not affected
by energy loss
The noise power analysis
shows interesting
characteristics of pCT
reconstruction algorithms
R Schulte, Proton Computed Tomography
Phase II: Clinical pCT
Large sensitive area (LLU/UCSC: 9 cm x 36 cm,
NIU: 27 cm x 36 cm)
Faster data acquisition (2 M protons/sec)
Faster image reconstruction (GP-GPU-based)
Integrated into a proton gantry
R Schulte, Proton Computed Tomography
Outlook: Applications for pCT
Higher planning accuracy/precision
Radiosurgery for vascular malformations,
pituitary adenomas, meningiomas, etc.
High-dose boosts to tumors near organs at
risk for damage
Creating lesions in defined locations for pain
Restricted beam angle choices
Paraspinal tumors
Craniospinal irradiation with sparing of
vertebral bodies in children
X-ray CT artifacts
R Schulte, Proton Computed Tomography
Where is the
Thank you!
R Schulte, Proton Computed Tomography
ResearchGate has not been able to resolve any citations for this publication.
ResearchGate has not been able to resolve any references for this publication.