Relevance of Monte Carlo modelling in quantitative imaging and
dosimetry in targeted therapy
PD Habib ZAIDI, Ph.D
Geneva University Hospital, Division of
Nuclear Medicine, CH-1211 Geneva, Switzerland
Email: habib.zaidi@hcuge.ch
Web: http:dmnu-pet5.hcuge.ch/
Abstract. There is no doubt that the area where
early Monte Carlo calculations in the field of medical
physics have been performed is internal dosimetry computations and modelling of
nuclear medicine instrumentation. The approach adopted by the Medical Internal
Radiation Dose (MIRD) committee for a so-called macrodose calculation make
extensive use of Monte Carlo calculations to derive specific absorbed fractions
for electron and photon sources uniformly distributed in organs of mathematical
phantoms. Nowadays, the applications
of the Monte Carlo method in nuclear medicine cover almost all topics including
detector modelling and imaging systems design, the assessment of image
quality and the quantitative accuracy of radionuclide imaging including development
and assessment of image
correction and reconstruction techniques, pharmacokinetic modelling, with an
increasing enthusiastic interest in exotic and exciting new applications such
as on-line PET monitoring of radiation therapy beams. This trend has continued
during the last decade and it is expected that Monte Carlo simulation techniques will find an
increasingly important role in the future of applications of quantitative
imaging and radiation dosimetry in targeted radiotherapy.
This lecture begins with an introduction to fundamental concepts
in image reconstruction and quantification. Impact of physical degrading factors
in radionuclides imaging including attenuation of photons and contribution from
photons scattered in the patient and partial volume effect on diagnostic
quality and quantitative accuracy of reconstructed images will be discussed. The
specific role of Monte Carlo simulations for the development and assessment of
quantitative imaging methodologies and patient-specific 3D dosimetry computations
for treatment planning in targeted therapy will be highlighted and illustrated
with examples from research carried out at the PET Instrumentation and
Neuroimaging Laboratory of Geneva University Hospital and other research labs
active in this field.
The combination of accurate models of the imaging
process allows the simulation of nuclear medicine data that are ever closer to
actual patient data. Monte Carlo
simulation techniques will find an increasingly important role in the future of
nuclear medicine in light of the further development of realistic computer
phantoms, the accurate modelling of projection data and computer hardware.
However, caution must be taken to avoid errors in the simulation process, and
verification via comparison with experimental and patient data is essential.