No abstract

Dose and range verification have become important tools to bring carbon ion therapy to a higher level of confidence in clinical applications. Positron emission tomography is among the most commonly used approaches for this purpose and relies on the creation of positron emitting nuclei in nuclear interactions of the primary ions with tissue. Predictions of these positron emitter distributions are usually obtained from time-consuming Monte Carlo simulations or measurements from previous treatment fractions, and their comparison to the current, measured image allows for treatment verification. Still, a direct comparison of planned and delivered dose would be highly desirable, since the dose is the quantity of interest in radiation therapy and its confirmation improves quality assurance in carbon ion therapy. In this work, we present a deconvolution approach to predict dose distributions from PET images in carbon ion therapy. Under the assumption that the one-dimensional PET distribution is described by a convolution of the depth dose distribution and a filter kernel, an evolutionary algorithm is introduced to perform the reverse step and predict the depth dose distribution from a measured PET distribution. Filter kernels are obtained from either a library or are created for any given situation on-the-fly, using predictions of the -decay and depth dose distributions, and the very same evolutionary algorithm. The applicability of this approach is demonstrated for monoenergetic and polyenergetic carbon ion irradiation of homogeneous and heterogeneous solid phantoms as well as a patient computed tomography image, using Monte Carlo simulated distributions and measured in-beam PET data. Carbon ion ranges are predicted within less than 0.5 mm and 1 mm deviation for simulated and measured distributions, respectively.

Comparative study of alternative Geant4 hadronic ion inelastic physics models Comparative study of alternative Geant4 hadronic ion inelastic physics models for prediction of positron-emitting radionuclide production in carbon and oxygen for prediction of positron-emitting radionuclide production in carbon and oxygen ion therapy ion therapy

This work presents an iterative method for the estimation of the absolute dose 24 distribution in patients undergoing carbon ion therapy, via analysis of the distribution 25 of positron annihilations resulting from the decay of positron-emitting fragments 26 created in the target volume. The proposed method relies on the decomposition of the total positron-annihilation distributions into profiles of the three principal 28 positron-emitting fragment species -11 C, 10 C and 15 O. A library of basis functions 29 is constructed by simulating a range of monoenergetic 12 C ion irradiations of a 30 homogeneous polymethyl methacrylate phantom and measuring the resulting one-31 dimensional positron-emitting fragment profiles and dose distributions. To estimate 32 the dose delivered during an arbitrary polyenergetic irradiation, a linear combination 33 of factors from the fragment profile library is iteratively fitted to the decomposed 34 positron annihilation profile acquired during the irradiation, and the resulting weights 35 combined with the corresponding monoenergetic dose profiles to estimate the total 36 dose distribution. A total variation regularisation term is incorporated into the fitting 37 process to suppress high-frequency noise. The method was evaluated with fourteen 38 different polyenergetic 12 C dose profiles in a polymethyl methacrylate target: one 39 which produces a flat biological dose, ten with randomised energy weighting factors, 40 and three with distinct dose maxima or minima within the spread-out Bragg peak 41 region. The proposed method is able to calculate the dose profile with mean relative 42 errors of 0.8%, 1.0% and 1.6% from the 11 C, 10 C, 15 O fragment profiles, respectively, 43 and estimate the position of the distal edge of the SOBP to within an average of 44 0.7 mm, 1.9 mm and 1.2 mm of its true location. 45 1. Introduction 46 Carbon ion therapy is a form of radiotherapy in which accelerated carbon ions are used 47 to deliver a therapeutic dose to the target volume [1, 2, 3]. This treatment modality 48 offers several advantages over photon therapy, such as a well-defined energy-dependent 49 depth of maximum dose shortly before the particles come to rest (known as the Bragg 50 peak), and a high relative biological effectiveness (RBE), particularly at the distal end 51 of the particle range [2, 4, 5]. The Bragg peak can be extended to deliver a uniform 52 dose over a depth range by superimposing monoenergetic beams with different energies 53 and fluences to form a polyenergetic beam, also known as a spread-out Bragg peak 54 (SOBP) [1]. However, anatomical changes, errors in patient positioning and errors in 55 the estimation of ion range may cause significant dose to be delivered outside the target 56 region due to the steep dose gradients between the target region and surrounding healthy 57 tissue [6]. 58 During carbon ion therapy, a variety of target and projectile fragments are produced 59 through nuclear inelastic collisions between ions in the beam and nuclei in the target 60 volum...

Purpose: Robotic radiosurgery offers the flexibility of a robotic arm to enable high conformity to the target and a steep dose gradient. However, treatment planning becomes a computationally challenging task as the search space for potential beam directions for dose delivery is arbitrarily large. We propose an approach based on deep learning to improve the search for treatment beams. Methods: In clinical practice, a set of candidate beams generated by a randomized heuristic forms the basis for treatment planning. We use a convolutional neural network to identify promising candidate beams. Using radiological features of the patient, we predict the influence of a candidate beam on the delivered dose individually and let this prediction guide the selection of candidate beams. Features are represented as projections of the organ structures which are relevant during planning. Solutions to the inverse planning problem are generated for random and CNN-predicted candidate beams. Results: The coverage increases from 95.35% to 97.67% for 6000 heuristically and CNN-generated candidate beams, respectively. Conversely, a similar coverage can be achieved for treatment plans with half the number of candidate beams. This results in a patient-dependent reduced averaged computation time of 20.28%-45.69%. The number of active treatment beams can be reduced by 11.35% on average, which reduces treatment time. Constraining the maximum number of candidate beams per beam node can further improve the average coverage by 0.75 percentage points for 6000 candidate beams. Conclusions: We show that deep learning based on radiological features can substantially improve treatment plan quality, reduce computation runtime, and treatment time compared to the heuristic approach used in clinics.