Statistical limitations in proton imaging

Energy painting: helium-beam radiography with thin detectors and multiple beam energies

Objective.Compact ion imaging systems based on thin detectors are a promising prospect for the clinical environment since they are easily integrated into the clinical workflow. Their measurement principle is based on energy deposition instead of the conventionally measured residual energy or range. Therefore, thin detectors are limited in the water-equivalent thickness range they can image with high precision. This article presents ourenergy paintingmethod, which has been developed to render high precision imaging with thin detectors feasible even for objects with larger, clinically relevant water-equivalent thickness (WET) ranges.Approach.A detection system exclusively based on pixelated silicon Timepix detectors was used at the Heidelberg ion-beam therapy center to track single helium ions and measure their energy deposition behind the imaged object. Calibration curves were established for five initial beam energies to relate the measured energy deposition to WET. They were evaluated regarding their accuracy, precision and temporal stability. Furthermore, a 60 mm × 12 mm region of a wedge phantom was imaged quantitatively exploiting the calibrated energies and five different mono-energetic images. These mono-energetic images were combined in a pixel-by-pixel manner by averaging the WET-data weighted according to their single-ion WET precision (SIWP) and the number of contributing ions.Main result.A quantitative helium-beam radiograph of the wedge phantom with an average SIWP of 1.82(5) % over the entire WET interval from 150 mm to 220 mm was obtained. Compared to the previously used methodology, the SIWP improved by a factor of 2.49 ± 0.16. The relative stopping power value of the wedge derived from the energy-painted image matches the result from range pullback measurements with a relative deviation of only 0.4 %.Significance.The proposed method overcomes the insufficient precision for wide WET ranges when employing detection systems with thin detectors. Applying this method is an important prerequisite for imaging of patients. Hence, it advances detection systems based on energy deposition measurements towards clinical implementation.

Technical note: development of a simulation framework, enabling the investigation of locally tuned single energy proton radiography

Range uncertainties remain a limitation for the confined dose distribution that proton therapy can offer. The uncertainty stems from the ambiguity when translating CT Hounsfield Units (HU) into proton stopping powers. Proton Radiography (PR) can be used to verify the proton range. Specifically, PR can be used as a quality-control tool for CBCT-based synthetic CTs. An essential part of the work illustrating the potential of PR has been conducted using multi-layer ionization chamber (MLIC) detectors and mono-energetic PR. Due to the dimensions of commercially available MLICs, clinical adoption is cumbersome. Here, we present a simulation framework exploring locally-tuned single energy (LTSE) proton radiography and corresponding potential compact PR detector designs. Based on a planning CT data set, the presented framework models the water equivalent thickness. Subsequently, it analyses the proton energies required to pass through the geometry within a defined ROI. In the final step, an LTSE PR is simulated using the MCsquare Monte Carlo code. In an anatomical head phantom, we illustrate that LTSE PR allows for a significantly shorter longitudinal dimension of MLICs. We compared PR simulations for two exemplary 30 × 30 mm2proton fields passing the phantom at a 90° angle at an anterior and a posterior location in an iso-centric setup. The longitudinal distance over which all spots per field range out is significantly reduced for LTSE PR compared to mono-energetic PR. In addition, we illustrate the difference in shape of integral depth dose (IDD) when using constrained PR energies. Finally, we demonstrate the accordance of simulated and experimentally acquired IDDs for an LTSE PR acquisition. As the next steps, the framework will be used to investigate the sensitivity of LTSE PR to various sources of errors. Furthermore, we will use the framework to systematically explore the dimensions of an optimized MLIC design for daily clinical use.

A denoising method based on deep learning for proton radiograph using energy resolved dose function

Objective.Proton radiograph has been broadly applied in proton radiotherapy which is affected by scattered protons which result in the lower spatial resolution of proton radiographs than that of x-ray images. Traditional image denoising method may lead to the change of water equivalent path length (WEPL) resulting in the lower WEPL measurement accuracy. In this study, we proposed a new denoising method of proton radiographs based on energy resolved dose function curves.Approach.Firstly, the corresponding relationship between the distortion of WEPL characteristic curve, and energy and proportion of scattered protons was established. Then, to improve the accuracy of proton radiographs, deep learning technique was used to remove scattered protons and correct deviated WEPL values. Experiments on a calibration phantom to prove the effectiveness and feasibility of this method were performed. In addition, an anthropomorphic head phantom was selected to demonstrate the clinical relevance of this technology and the denoising effect was analyzed.Main results.The curves of WEPL profiles of proton radiographs became smoother and deviated WEPL values were corrected. For the calibration phantom proton radiograph, the average absolute error of WEPL values decreased from 2.23 to 1.72, the mean percentage difference of all materials of relative stopping power decreased from 1.24 to 0.39, and the average relative WEPL corrected due to the denoising process was 1.06%. In addition, WEPL values correcting were also observed on the proton radiograph for anthropomorphic head phantom due to this denoising process.Significance.The experiments showed that this new method was effective for proton radiograph denoising and had greater advantages than end-to-end image denoising methods, laying the foundation for the implementation of precise proton radiotherapy.

Extension of the open-source TIGRE toolbox for proton imaging

Proton irradiation is a well-established method to treat deep-seated tumors in radio oncology. Usually, an X-ray computed tomography (CT) scan is used for treatment planning. Since proton therapy is based on the precise knowledge of the stopping power describing the energy loss of protons in the patient tissues, the Hounsfield units of the planning CT have to be converted. This conversion introduces range errors in the treatment plan, which could be reduced, if the stopping power values were extracted directly from an image obtained using protons instead of X-rays. Since protons are affected by multiple Coulomb scattering, reconstruction of the 3D stopping power map results in limited image quality if the curved proton path is not considered. This work presents a substantial code extension of the open-source toolbox TIGRE for proton CT (pCT) image reconstruction based on proton radiographs including a curved proton path estimate. The code extension and the reconstruction algorithms are GPU-based, allowing to achieve reconstruction results within minutes. The performance of the pCT code extension was tested with Monte Carlo simulated data using three phantoms (Catphan® high resolution and sensitometry modules and a CIRS patient phantom). In the simulations, ideal and non-ideal conditions for a pCT setup were assumed. The obtained mean absolute percentage error was found to be below 1% and up to 8 lp/cm could be resolved using an idealized setup. These findings demonstrate that the presented code extension to the TIGRE toolbox offers the possibility for other research groups to use a fast and accurate open-source pCT reconstruction.

Experimental helium-beam radiography with a high-energy beam: Water-equivalent thickness calibration and first image-quality results

PURPOSE

A clinical implementation of ion-beam radiography (iRad) is envisaged to provide a method for on-couch verification of ion-beam treatment plans. The aim of this work is to introduce and evaluate a method for quantitative water-equivalent thickness (WET) measurements for a specific helium-ion imaging system for WETs that are relevant for imaging thicker body parts in the future.

METHODS

Helium-beam radiographs (αRads) are measured at the Heidelberg Ion-beam Therapy Center (HIT) with an initial beam energy of 239.5 MeV/ u. An imaging system based on three pairs of thin silicon pixel detectors is used for ion path reconstruction and measuring the energy deposition (dE) of each particle behind the object to be imaged. The dE behind homogeneous plastic blocks is related to their well-known WETs between 280.6mm and 312.6 mm with a calibration curve that is created by fitting the measured data points. The quality of the quantitative WET measurements is determined by the uncertainty of the measured WET of a single ion (single-ion WET precision) and the deviation of a measured WET value to the well-known WET (WET accuracy). Subsequently, the fitted calibration curve is applied to an energy deposition radiograph of a phantom with a complex geometry. The spatial resolution (modulation transfer function at 10% (MTF_10% )) and WET accuracy (mean absolute percentage difference (MAPD)) of the WET map, are determined.

RESULTS

In the optimal imaging WET-range from ∼  280 mm to 300 mm, the fitted calibration curve reached a mean single-ion WET precision of 1.55 ± 0.00%. Applying the calibration to an ion radiograph (iRad) of a more complex WET distribution, the spatial resolution was determined to be MTF_10% = 0.49 ± 0.03 lp/mm and the WET accuracy was assessed as MAPD to 0.21%.

CONCLUSIONS

Using a beam energy of 239.5MeV/ u and the proposed calibration procedure, quantitative αRads of WETs between ∼  280mm to 300 mm can be measured and show high potential for clinical use. The proposed approach with the resulting image qualities encourages further investigation towards the clinical application of helium-beam radiography. This article is protected by copyright. All rights reserved.

Improved accuracy of relative electron density and proton stopping power ratio through CycleGAN machine learning

Objective. Kilovoltage computed tomography (kVCT) is the cornerstone of radiotherapy treatment planning for delineating tissues and towards dose calculation. For the former, kVCT provides excellent contrast and signal-to-noise ratio. For the latter, kVCT may have greater uncertainty in determining relative electron density ( ρ e ) and proton stopping power ratio (SPR). Conversely, megavoltage CT (MVCT) may result in superior dose calculation accuracy. The purpose of this work was to convert kVCT HU to MVCT HU using deep learning to obtain higher accuracy ρ e and SPR. Approach. Tissue-mimicking phantoms were created to compare kVCT- and MVCT-determined ρ e and SPR to physical measurements. Using 100 head-and-neck datasets, an unpaired deep learning model was trained to learn the relationship between kVCTs and MVCTs, creating synthetic MVCTs (sMVCTs). Similarity metrics were calculated between kVCTs, sMVCTs, and MVCTs in 20 test datasets. An anthropomorphic head phantom containing bone-mimicking material with known composition was scanned to provide an independent determination of ρ e and SPR accuracy by sMVCT. Main results. In tissue-mimicking bone, ρ e errors were 2.20% versus 0.19% and SPR errors were 4.38% versus 0.22%, for kVCT versus MVCT, respectively. Compared to MVCT, in vivo mean difference (MD) values were 11 and 327 HU for kVCT and 2 and 3 HU for sMVCT in soft tissue and bone, respectively. ρ e MD decreased from 1.3% to 0.35% in soft tissue and 2.9% to 0.13% in bone, for kVCT and sMVCT, respectively. SPR MD decreased from 1.8% to 0.24% in soft tissue and 6.8% to 0.16% in bone, for kVCT and sMVCT, respectively. Relative to physical measurements, ρ e and SPR error in anthropomorphic bone decreased from 7.50% and 7.48% for kVCT to <1% for both MVCT and sMVCT. Significance. Deep learning can be used to map kVCT to sMVCT, suggesting higher accuracy ρ e and SPR is achievable with sMVCT versus kVCT.

Experimental comparison of photon versus particle computed tomography to predict tissue relative stopping powers

PURPOSE

Measurements comparing relative stopping power (RSP) accuracy of state-of-the-art systems representing single-energy and dual-energy computed tomography (SECT/DECT) with proton CT (pCT) and helium CT (HeCT) in biological tissue samples.

METHODS

We used 16 porcine and bovine samples of various tissue types and water, covering an RSP range from 0.90 ± 0.06 to 1.78 ± 0.05. Samples were packed and sealed into 3D-printed cylinders ( d = 2  cm, h = 5  cm) and inserted into an in-house designed cylindrical polymethyl methacrylate (PMMA) phantom ( d = 10  cm, h = 10  cm). We scanned the phantom in a commercial SECT and DECT (120 kV; 100  and 140 kV/Sn (tin-filtered)); and acquired pCT and HeCT ( E ∼ 200  MeV/u, 2 ∘ steps, ∼ 6.2 × 10 6 (p)/ ∼ 2.3 × 10 6 (He) particles/projection) with a particle imaging prototype. RSP maps were calculated from SECT/DECT using stoichiometric methods and from pCT/HeCT using the DROP-TVS algorithm. We estimated the average RSP of each tissue per modality in cylindrical volumes of interest and compared it to ground truth RSP taken from peak-detection measurements.

RESULTS

Throughout all samples, we observe the following root-mean-squared RSP prediction errors ± combined uncertainty from reference measurement and imaging: SECT 3.10 ± 2.88%, DECT 0.75 ± 2.80%, pCT 1.19 ± 2.81%, and HeCT 0.78 ± 2.81%. The largest mean errors ± combined uncertainty per modality are SECT 8.22 ± 2.79% in cortical bone, DECT 1.74 ± 2.00% in back fat, pCT 1.80 ± 4.27% in bone marrow, and HeCT 1.37 ± 4.25% in bone marrow. Ring artifacts were observed in both pCT and HeCT reconstructions, imposing a systematic shift to predicted RSPs.

CONCLUSION

Comparing state-of-the-art SECT/DECT technology and a pCT/HeCT prototype, DECT provided the most accurate RSP prediction, closely followed by particle imaging. The novel modalities pCT and HeCT have the potential to further improve on RSP accuracies with work focusing on the origin and correction of ring artifacts. Future work will study accuracy of proton treatment plans using RSP maps from investigated imaging modalities.

Image quality evaluation of projection- and depth dose-based approaches to integrating proton radiography using a monolithic scintillator detector

The purpose of this study is to compare the image quality of an integrating proton radiography (PR) system, composed of a monolithic scintillator and two digital cameras, using integral lateral-dose and integral depth-dose image reconstruction techniques. Monte Carlo simulations were used to obtain the energy deposition in a 3D monolithic scintillator detector (30 × 30 × 30 cm³poly vinyl toluene organic scintillator) to create radiographs of various phantoms-a slanted aluminum cube for spatial resolution analysis and a Las Vegas phantom for contrast analysis. The light emission of the scintillator was corrected using Birks scintillation model. We compared two integrating PR methods and the expected results from an idealized proton tracking radiography system. Four different image reconstruction methods were utilized in this study: integral scintillation light projected from the beams-eye view, depth-dose based reconstruction methods both with and without optimization, and single particle tracking PR was used for reference data. Results showed that heterogeneity artifact due to medium-interface mismatch was identified from the Las Vegas phantom simulated in air. Spatial resolution was found to be highest for single-event reconstruction. Contrast levels, ranked from best to worst, were found to correspond to particle tracking, optimized depth-dose, depth-dose, and projection-based image reconstructions. The image quality of a monolithic scintillator integrating PR system was sufficient to warrant further exploration. These results show promise for potential clinical use as radiographic techniques for visualizing internal patient anatomy during proton radiotherapy.

Statistical limitations in ion imaging

In this study, we investigated the capacity of various ion beams available for radiotherapy to produce high quality relative stopping power map acquired from energy-loss measurements. The image quality metrics chosen to compare the different ions were signal-to-noise ratio (SNR) as a function of dose and spatial resolution. Geant4 Monte Carlo simulations were performed for: hydrogen, helium, lithium, boron and carbon ion beams crossing a 20 cm diameter water phantom to determine SNR and spatial resolution. It has been found that protons possess a significantly larger SNR when compared with other ions at a fixed range (up to 36% higher than helium) due to the proton nuclear stability and low dose per primary. However, it also yields the lowest spatial resolution against all other ions, with a resolution lowered by a factor 4 compared to that of carbon imaging, for a beam with the same initial range. When comparing for a fixed spatial resolution of 10 lp cm⁻¹, carbon ions produce the highest image quality metrics with proton ions producing the lowest. In conclusion, it has been found that no ion can maximize all image quality metrics simultaneously and that a choice must be made between spatial resolution, SNR, and dose.

Fluence-modulated proton CT optimized with patient-specific dose and variance objectives for proton dose calculation

Particle therapy treatment planning requires accurate volumetric maps of the relative stopping power, which can directly be acquired using proton computed tomography (pCT). With fluence-modulated pCT (FMpCT) imaging fluence is concentrated in a region-of-interest (ROI), which can be the vicinity of the treatment beam path, and imaging dose is reduced elsewhere. In this work we present a novel optimization algorithm for FMpCT which, for the first time, calculates modulated imaging fluences for joint imaging dose and image variance objectives. Thereby, image quality is maintained in the ROI to ensure accurate calculations of the treatment dose, and imaging dose is minimized outside the ROI with stronger minimization penalties given to imaging organs-at-risk. The optimization requires an initial scan at uniform fluence or a previous x-ray CT scan. We simulated and optimized FMpCT images for three pediatric patients with tumors in the head region. We verified that the target image variance inside the ROI was achieved and demonstrated imaging dose reductions outside of the ROI of 74% on average, reducing the imaging dose from 1.2 to 0.3 mGy. Such dose savings are expected to be relevant compared to the therapeutic dose outside of the treatment field. Treatment doses were re-calculated on the FMpCT images and compared to treatment doses re-recalculated on uniform fluence pCT scans using a 1% criterion. Passing rates were above 98.3% for all patients. Passing rates comparing FMpCT treatment doses to the ground truth treatment dose were above 88.5% for all patients. Evaluation of the proton range with a 1 mm criterion resulted in passing rates above 97.5% (FMpCT/pCT) and 95.3% (FMpCT/ground truth). Jointly optimized fluence-modulated pCT images can be used for proton dose calculation maintaining the full dosimetric accuracy of pCT but reducing the required imaging dose considerably by three quarters. This may allow for daily imaging during particle therapy ensuring a safe and accurate delivery of the therapeutic dose and avoiding excess dose from imaging.

The role of Monte Carlo simulation in understanding the performance of proton computed tomography

Proton computed tomography (pCT) is a promising tomographic imaging modality allowing direct reconstruction of proton relative stopping power (RSP) required for proton therapy dose calculation. In this review article, we aim at highlighting the role of Monte Carlo (MC) simulation in pCT studies. After describing the requirements for performing proton computed tomography and the various pCT scanners actively used in recent research projects, we present an overview of available MC simulation platforms. The use of MC simulations in the scope of investigations of image reconstruction, and for the evaluation of optimal RSP accuracy, precision and spatial resolution omitting detector effects is then described. In the final sections of the review article, we present specific applications of realistic MC simulations of an existing pCT scanner prototype, which we describe in detail.