Using dual-energy x-ray imaging to enhance automated lung tumor tracking during real-time adaptive radiotherapy

Deep learning-based target decomposition for markerless lung tumor tracking in radiotherapy

BACKGROUND

In radiotherapy, real-time tumor tracking can verify tumor position during beam delivery, guide the radiation beam to target the tumor, and reduce the chance of a geometric miss. Markerless kV x-ray image-based tumor tracking is challenging due to the low tumor visibility caused by tumor-obscuring structures. Developing a new method to enhance tumor visibility for real-time tumor tracking is essential.

PURPOSE

To introduce a novel method for markerless kV image-based tracking of lung tumors via deep learning-based target decomposition.

METHODS

We utilized a conditional Generative Adversarial Network (cGAN), known as Pix2Pix, to build a patient-specific model and generate the synthetic decomposed target image (sDTI) to enhance tumor visibility on the real-time kV projection images acquired by the onboard kV imager equipped on modern linear accelerators. We used 4DCT simulation images to generate the digitally reconstructed radiograph (DRR) and DTI image pairs for model training. We augmented the training dataset by randomly shifting the 4DCT in the superior-inferior, anterior-posterior, and left-right directions during the DRR and DTI generation process. We performed real-time 2D tumor tracking via template matching between the DTI generated from the CT simulation and the sDTI generated from the real-time kV projection images. We validated the proposed method using nine patients' datasets with implanted beacons near the tumor.

RESULTS

The sDTI can effectively improve the image contrast around the lung tumors on the kV projection images for the nine patients. With the beacon motion as ground truth, the tracking errors were on average 0.8 ± 0.7 mm in the superior-inferior (SI) direction and 0.9 ± 0.8 mm in the in-plane left-right (IPLR) direction. The percentage of successful tracking, defined as a tracking error less than 2 mm in the SI direction, is 92.2% on the 4312 tested images. The patient-specific model took approximately 12 h to train. During testing, it took approximately 35 ms to generate one sDTI, and 13 ms to perform the tumor tracking using template matching.

CONCLUSIONS

Our method offers the potential solution for nearly real-time markerless lung tumor tracking. It achieved a high level of accuracy and an impressive tracking rate. Further development of 3D lung tumor tracking is warranted.

Deep match: A zero-shot framework for improved fiducial-free respiratory motion tracking

BACKGROUND AND PURPOSE

Motion management is essential to reduce normal tissue exposure and maintain adequate tumor dose in lung stereotactic body radiation therapy (SBRT). Lung SBRT using an articulated robotic arm allows dynamic tracking during radiation dose delivery. Two stereoscopic X-ray tracking modes are available - fiducial-based and fiducial-free tracking. Although X-ray detection of implanted fiducials is robust, the implantation procedure is invasive and inapplicable to some patients and tumor locations. Fiducial-free tracking relies on tumor contrast, which challenges the existing tracking algorithms for small (e.g., <15 mm) and/or tumors obscured by overlapping anatomies. To markedly improve the performance of fiducial-free tracking, we proposed a deep learning-based template matching algorithm - Deep Match.

METHOD

Deep Match consists of four self-definable stages - training-free feature extractor, similarity measurements for location proposal, local refinements, and uncertainty level prediction for constructing a more trustworthy and versatile pipeline. Deep Match was validated on a 10 (38 fractions; 2661 images) patient cohort whose lung tumor was trackable on one X-ray view, while the second view did not offer sufficient conspicuity for tumor tracking using existing methods. The patient cohort was stratified into subgroups based on tumor sizes (<10 mm, 10-15 mm, and >15 mm) and tumor locations (with/without thoracic anatomy overlapping).

RESULTS

On X-ray views that conventional methods failed to track the lung tumor, Deep Match achieved robust performance as evidenced by >80 % 3 mm-Hit (detection within 3 mm superior/inferior margin from ground truth) for 70 % of patients and <3 mm superior/inferior distance (SID) ∼1 mm standard deviation for all the patients.

CONCLUSION

Deep Match is a zero-shot learning network that explores the intrinsic neural network benefits without training on patient data. With Deep Match, fiducial-free tracking can be extended to more patients with small tumors and with tumors obscured by overlapping anatomy.

Enhancing the target visibility with synthetic target specific digitally reconstructed radiograph for intrafraction motion monitoring: A proof-of-concept study

BACKGROUND

Intrafraction motion monitoring in External Beam Radiation Therapy (EBRT) is usually accomplished by establishing a correlation between the tumor and the surrogates such as an external infrared reflector, implanted fiducial markers, or patient skin surface. These techniques either have unstable surrogate-tumor correlation or are invasive. Markerless real-time onboard imaging is a noninvasive alternative that directly images the target motion. However, the low target visibility due to overlapping tissues along the X-ray projection path makes tumor tracking challenging.

PURPOSE

To enhance the target visibility in projection images, a patient-specific model was trained to synthesize the Target Specific Digitally Reconstructed Radiograph (TS-DRR).

METHODS

Patient-specific models were built using a conditional Generative Adversarial Network (cGAN) to map the onboard projection images to TS-DRR. The standard Pix2Pix network was adopted as our cGAN model. We synthesized the TS-DRR based on the onboard projection images using phantom and patient studies for spine tumors and lung tumors. Using previously acquired CT images, we generated DRR and its corresponding TS-DRR to train the network. For data augmentation, random translations were applied to the CT volume when generating the training images. For the spine, separate models were trained for an anthropomorphic phantom and a patient treated with paraspinal stereotactic body radiation therapy (SBRT). For lung, separate models were trained for a phantom with a spherical tumor insert and a patient treated with free-breathing SBRT. The models were tested using Intrafraction Review Images (IMR) for the spine and CBCT projection images for the lung. The performance of the models was validated using phantom studies with known couch shifts for the spine and known tumor deformation for the lung.

RESULTS

Both the patient and phantom studies showed that the proposed method can effectively enhance the target visibility of the projection images by mapping them into synthetic TS-DRR (sTS-DRR). For the spine phantom with known shifts of 1 mm, 2 mm, 3 mm, and 4 mm, the absolute mean errors for tumor tracking were 0.11 ± 0.05 mm in the x direction and 0.25 ± 0.08 mm in the y direction. For the lung phantom with known tumor motion of 1.8 mm, 5.8 mm, and 9 mm superiorly, the absolute mean errors for the registration between the sTS-DRR and ground truth are 0.1 ± 0.3 mm in both the x and y directions. Compared to the projection images, the sTS-DRR has increased the image correlation with the ground truth by around 83% and increased the structural similarity index measure with the ground truth by around 75% for the lung phantom.

CONCLUSIONS

The sTS-DRR can greatly enhance the target visibility in the onboard projection images for both the spine and lung tumors. The proposed method could be used to improve the markerless tumor tracking accuracy for EBRT.

Monte Carlo model of a prototype flat-panel detector for multi-energy applications in radiotherapy

BACKGROUND

The incorporation of multi-energy capabilities into radiotherapy flat-panel detectors offers advantages including enhanced soft tissue visualization by reduction of signal from overlapping anatomy such as bone in 2D image projections; creation of virtual monoenergetic images for 3D contrast enhancement, metal artefact reduction and direct acquisition of relative electron density. A novel dual-layer on-board imager offering dual energy processing capabilities is being designed. As opposed to other dual-energy implementation techniques which require separate acquisition with two different x-ray spectra, the dual-layer detector design enables simultaneous acquisition of high and low energy images with a single exposure. A computational framework is required to optimize the design parameters and evaluate detector performance for specific clinical applications.

PURPOSE

In this study, we report on the development of a Monte Carlo (MC) model of the imager including model validation.

METHODS

The stack-up of the dual-layer imager (DLI) was implemented in GEANT4 Application for Tomographic Emission (GATE). The DLI model has an active area of 43×43 cm2 , with top and bottom Cesium Iodide (CsI) scintillators of 600 and 800 μm thickness, respectively. Measurement of spatial resolution and imaging of dedicated multi-material dual-energy (DE) phantoms were used to validate the model. The modulation transfer function (MTF) of the detector was calculated for a 120 kVp x-ray spectrum using a 0.5 mm thick tantalum edge rotated by 2.5o . For imaging validation, the DE phantom was imaged using a 140 kVp x-ray spectrum. For both validation simulations, corresponding measurements were done using an initial prototype of the imager. Agreement between simulations and measurement was assessed using normalized root mean square error (NRMSE) and 1D profile difference for the MTF and phantom images respectively. Further comparison between measurement and simulation was made using virtual monoenergetic images (VMIs) generated from basis material images derived using precomputed look-up tables.

RESULTS

The MTF of the bottom layer of the dual-layer model shows values decreasing more quickly with spatial frequency, compared to the top layer, due to the thicker bottom scintillator thickness and scatter from the top layer. A comparison with measurement shows NRMSE of 0.013 and 0.015 as well as identical MTF50 of 0.8 mm1 and 1.0 mm1 for the top and bottom layer respectively. For the DE imaging of the DE-phantom, although a maximum deviation of 3.3% is observed for the 10 mm aluminum and Teflon inserts at the top layer, the agreement for all other inserts is less than 2.2% of the measured value at both layers. Material decomposition of simulated scatter-free DE images gives an average accuracy in PMMA and aluminum composition of 4.9% and 10.3% for 11-30 mm PMMA and 1-10 mm aluminum objects respectively. A comparison of decomposed values using scatter containing measured and simulated DE images shows good agreement within statistical uncertainty.

CONCLUSION

Validation using both MTF and phantom imaging shows good agreement between simulation and measurements. With the present configuration of the digital prototype, the model can generate material decomposed images and virtual monoenergetic images.

Spatial distribution of the imaging dose and characterization of the scatter radiation contribution in CyberKnife radiosurgery

PURPOSE

The imaging dose for intra- and extra-cranial CyberKnife radiosurgery applications was calculated and the scattered radiation reaching the digital detectors was quantified and analyzed with regard to its origin.

METHODS

The image guidance subsystem of the CyberKnife was modeled based on vendor-provided information. The emitted X-ray energy spectrum for 120 kV was estimated using the SpekPy software tool. Monte Carlo (MC) image acquisition simulations were performed to calculate the total, primary and scattered photon fluences reaching each detector as a function of the imaged object dimensions. MC calculations of the imaging dose were performed for intra- and extra-cranial applications assuming 120 kV and 10 mAs acquisition settings.

RESULTS

The amount of scattered radiation reaching each detector was found to depend on the dimensions of the imaged anatomical region, contributing more than 40 % to the total photon fluence for regions more than 20 cm thick. More than 20 % of this scattered radiation originates from the contralateral imaging field. A maximum organ dose of 1.5 mGy at the nasal bones and an average dose of 0.37 mGy to the eye lenses per image pair acquisition was calculated for head applications. An entrance imaging dose of 0.4 mGy was calculated for extracranial applications.

CONCLUSIONS

Scattered radiation reaching each detector in the skull and spine tracking applications can be reduced by acquiring the pair of radiographs sequentially instead of simultaneously. A dose of 3.7 cGy to the eye lenses is estimated assuming 100 image pair exposures required for treatment completion.

A novel bone suppression algorithm in intensity‐based 2D/3D image registration for real‐time tumour motion monitoring: development and phantom‐based validation

Background

Real‐time tumor motion monitoring (TMM) is a crucial process for intra‐fractional respiration management in lung cancer radiotherapy. Since the tumor can be partly or fully located behind the ribs, the TMM is challenging.

Purpose

The aim of this work was to develop a bone suppression (BS) algorithm designed for real‐time 2D/3D marker‐less TMM to increase the visibility of the tumor when overlapping with bony structures and consequently to improve the accuracy of TMM.

Method

A BS method was implemented in the in‐house developed software for ultrafast intensity‐based 2D/3D tumor registration (Fast Image‐based Registration [FIRE]). The method operates on both, digitally reconstructed radiograph (DRR) and intra‐fractional X‐ray images. The bony structures are derived from computed tomography data by thresholding during ray‐casting, and the resulting bone DRR is subtracted from intra‐fractional X‐ray images to obtain a soft‐tissue‐only image for subsequent tumor registration. The accuracy of TMM utilizing BS was evaluated within a retrospective phantom study with nine different 3D‐printed tumor phantoms placed in the in‐house developed Advanced Radiation DOSimetry (ARDOS) breathing phantom. A 24 mm craniocaudal tumor motion, including rib eclipses, was simulated, and X‐ray images were acquired on the Elekta Versa HD Linac in the lateral and posterior–anterior directions. An error assessment for BS images was evaluated with respect to the ground truth tumor position.

Results

A total error (root mean square error) of 0.87 ± 0.23 mm and 1.03 ± 0.26 mm was found for posterior–anterior and lateral imaging; the mean time for BS was 8.03 ± 1.54 ms. Without utilizing BS, TMM failed in all X‐ray images since the registration algorithm focused on the rib position due to the predominant intensity of this tissue within DRR and X‐ray images.

Conclusion

The BS algorithm developed and implemented improved the accuracy, robustness, and stability of real‐time TMM in lung cancer in a phantom study, even in the case of rib interlude where normal tumor registration fails.

Adaptive dual-energy algorithm based on pre-calibrated weighting factors for chest radiography

Objective. To develop a dual-energy (DE) algorithm with spatially varying weighting factors for material selection and noise suppression. Approach. Calibration step-phantoms, with overlapping slabs of solid water and bone with different thicknesses, were used to obtain the pre-calibrated material selection and noise reduction weighting factors. The Material selection weighting factors were calculated by finding a zero of contrast-to-noise-ratio (CNR) between regions with two overlapping materials and regions of only target material, while noise suppression weighting factors were determined by maximizing signal-to-noise ratio for overlapping regions. The pre-calibrated weighting factors were fitted with low and high energy radiograph of two Rando phantoms to create maps of material selection and noise suppression weighting factors, which used with DE algorithm and anti-correlated noise reduction (ACNR) algorithm to generate DE images. Three different implementations, including two different sizes of Rando phantoms and two different orientations (oblique and anterior-posterior), were investigated. Soft-tissue and bone only images of Rando phantoms were obtained with five combinations of DE algorithms and CNR, contrast, and noise values of selected regions of interest were compared to evaluate the performance of the novel method: simple log subtraction (SLS), SLS with uniform ACNR, adaptive DE (aDE), aDE with uniform ACNR, and aDE and adaptive ACNR (aACNR). Main results. Compared to SLS, the aDE algorithm demonstrated improved image quality in all three orientations. CNR increased with better contrast for both soft-tissue and bone images. Implementation of aACNR algorithm resulted in further reduction of image noise and improvements in CNR at the cost of contrast. However, aACNR algorithm showed better contrast compared to ACNR method. Significance. A novel DE algorithm was proposed, which showed improved material selection and noise suppression as compared to the conventional DE techniques and can be easily implemented in a clinical environment for real-time DE image generation.

The markerless lung target tracking AAPM Grand Challenge (MATCH) results

PURPOSE

Lung stereotactic ablative body radiotherapy (SABR) is a radiation therapy success story with level 1 evidence demonstrating its efficacy. To provide real-time respiratory motion management for lung SABR, several commercial and preclinical markerless lung target tracking (MLTT) approaches have been developed. However, these approaches have yet to be benchmarked using a common measurement methodology. This knowledge gap motivated the MArkerless lung target Tracking CHallenge (MATCH). The aim was to localize lung targets accurately and precisely in a retrospective in silico study and a prospective experimental study.

METHODS

MATCH was an American Association of Physicists in Medicine sponsored Grand Challenge. Common materials for the in silico and experimental studies were the experiment setup including an anthropomorphic thorax phantom with two targets within the lungs, and a lung SABR planning protocol. The phantom was moved rigidly with patient-measured lung target motion traces, which also acted as ground truth motion. In the retrospective in silico study a volumetric modulated arc therapy treatment was simulated and a dataset consisting of treatment planning data and intra-treatment kilovoltage (kV) and megavoltage (MV) images for four blinded lung motion traces was provided to the participants. The participants used their MLTT approach to localize the moving target based on the dataset. In the experimental study, the participants received the phantom experiment setup and five patient-measured lung motion traces. The participants used their MLTT approach to localize the moving target during an experimental SABR phantom treatment. The challenge was open to any participant, and participants could complete either one or both parts of the challenge. For both the in silico and experimental studies the MLTT results were analyzed and ranked using the prospectively defined metric of the percentage of the tracked target position being within 2 mm of the ground truth.

RESULTS

A total of 30 institutions registered and 15 result submissions were received, four for the in silico study and 11 for the experimental study. The participating MLTT approaches were: Accuray CyberKnife (2), Accuray Radixact (2), BrainLab Vero, C-RAD, and preclinical MLTT (5) on a conventional linear accelerator (Varian TrueBeam). For the in silico study the percentage of the 3D tracking error within 2 mm ranged from 50% to 92%. For the experimental study, the percentage of the 3D tracking error within 2 mm ranged from 39% to 96%.

CONCLUSIONS

A common methodology for measuring the accuracy of MLTT approaches has been developed and used to benchmark preclinical and commercial approaches retrospectively and prospectively. Several MLTT approaches were able to track the target with sub-millimeter accuracy and precision. The study outcome paves the way for broader clinical implementation of MLTT. MATCH is live, with datasets and analysis software being available online at https://www.aapm.org/GrandChallenge/MATCH/ to support future research.

Noise-robust breathing-phase estimation on marker-free, ultra low dose X-ray projections for real-time tumor localization via surrogate structures

PURPOSE

This paper presents a novel strategy for feature-based breathing-phase estimation on ultra low-dose X-ray projections for tumor motion control in radiation therapy.

METHODS

Coarse-scaled Curvelet coefficients are identified as motion sensitive but noise-robust features for this purpose. For feature-based breathing-phase estimation, an ensemble strategy with two classifiers is used. This consensus-based estimation substantially increases tracking reliability by rejection of false positives. The algorithm is evaluated on both synthetic and measured phantom data: Monte Carlo simulated ultra low dose projections for a C-arm X-ray and on the basis of 4D-chest-CTs of eight patients on one hand side and real measurements based on a motion phantom.

RESULTS

To achieve an accuracy of breathing-phase estimation of more than 95% a fluence between 20 and 400 photons per pixel (open field) is required depending on the patient. Furthermore, the algorithm is evaluated on real ultra low dose projections from an XVI R5.0 system (Elekta AB, Stockholm, Sweden) using an additional lead filter to reduce fluence. The classifiers-consensus-based-gating method estimated the correct position of the test projections in all test cases at a fluence of ∼180 photons per pixel and 92% at a fluence of ∼40 photons per pixel. The deposited dose to patient per image is in the range of nGy.

CONCLUSIONS

A novel method is presented for estimation of breathing-phases for real-time tumor localization at ultra low dose both on a simulation and a phantom basis. Its accuracy is comparable to state of the art X-ray based algorithms while the released dose to patients is reduced by two to three orders of magnitude compared to conventional template-based approaches. This allows for continuous motion control during irradiation without the need of external markers.

Comprehensive characterization of ExacTrac stereoscopic image guidance system using Monte Carlo and Spektr simulations

The purpose of this work is to develop accurate computational methods to comprehensively characterize and model the clinical ExacTrac imaging system, which is used as an image guidance system for stereotactic treatment applications. The Spektr toolkit was utilized to simulate the spectral and imaging characterization of the system. Since Spektr only simulates the primary beam (ignoring scatter), a full model of ExacTrac was also developed in Monte Carlo (MC) to characterize the imaging system. To ensure proper performance of both simulation models, Spektr and MC data were compared to the measured spectral and half value layers (HVLs) values. To validate the simulation results, x-ray spectra of the ExacTrac system were measured for various tube potentials using a CdTe spectrometer with multiple added narrow collimators. The raw spectra were calibrated using a ⁵⁷Co source and corrected for the escape peaks and detector efficiency. HVLs in mm of Al for various energies were measured using a calibrated RaySafe detector. Spektr and MC HVLs were calculated and compared to the measured values. The patient surface dose was calculated for different clinical imaging protocols from the measured air kerma and HVL values following the TG-61 methodology. The x-ray focal spot was measured by slanted edge technique using gafchromic films. ExacTrac imaging system beam profiles were simulated for various energies by MC simulation and the results were benchmarked by experimentally acquired beam profiles using gafchromic films. The effect of 6D IGRT treatment couch on beam hardening, dynamic range of the flat panel detector and scatter effect were determined using both Spektr simulation and experimental measurements. The measured and simulated spectra (of both MC and Spektr) for various kVps were compared and agreed within acceptable error. As another validation, the measured HVLs agreed with the Spektr and MC simulated HVLs on average within 1.0% for all kVps. The maximum and minimum patient surface doses were found to be 1.06 mGy for shoulder (high) and 0.051 mGy for cranial (low) imaging protocols, respectively. The MC simulated beam profiles were well matched with experimental results and replicated the penumbral slopes, the heel effect, and out-of-field regions. Dynamic range of detector (in terms of air kerma at detector surface) was found to be in the range of [6.1 × 10^-6, 5.3 × 10^-3] mGy. Accurate MC and Spektr models of the ExacTrac image guidance system were successfully developed and benchmarked via experimental validation. While patient surface dose for available imaging protocols were reported in this study, the established MC model may be used to obtain 3D imaging dose distribution for real patient geometries.

Adaptive noise reduction for dual-energy x-ray imaging based on spatial variations in beam attenuation

PURPOSE

The main goal of this work is to improve the previously proposed patient-specific pixel-based dual-energy (PP-DE) algorithm by developing an adaptive anti-correlated noise reduction (ACNR) method, resulting in reduced image noise.

METHODS

Theoretical models of contrast-to-noise (CNR) and signal-to-noise (SNR) ratio were developed as functions of weighting factors for DE bone ω _Bn or soft tissue ω _ST cancellation. These analytical expressions describe CNR and SNR properties of dual-energy (DE) images, obtained with both simple log subtraction (SLS) and ACNR algorithms, and allow for a direct comparison between experimental and theoretical results. The theoretical models demonstrate the importance of ACNR weighting factor (ω _A ) optimization leading to the maximization of the SNR of the final image. A step phantom was constructed, which consisted of overlapping slabs of solid water (0-30 cm) and bone-mimicking material (0-6 cm), resulting in a total of 7 × 7 regions. High-energy (HE) and low-energy (LE) images were acquired at 140 kVp and 60 kVp with a clinical ExacTrac imaging system. The CNR and SNR were obtained for the DE images as functions of ω _Bn,ST and noise reduction weighting factor ω _A for different combinations of thicknesses. Weighting factors for bone cancellation were optimized for each region of interest (ROI) by finding zeros of CNR function for DE images between soft tissue only and soft tissue plus bone regions (and vice versa for soft tissue cancellation). The weighting factor for the ACNR algorithm ω _A was then optimized by maximizing the SNR function for each ROI. HE and LE images for an anthropomorphic Rando phantom were obtained with the same acquisition parameters as for the step phantom. DE images for bone only and soft tissue only were obtained with three algorithms: SLS and PP-DE with conventional ACNR (uniform ω _A ), and PP-DE with adaptive ACNR (region-varying ω _A ). Weighting factor maps for PP-DE and adaptive ACNR methods were obtained for Rando phantom geometry (which was determined from its CT scans) by interpolation (or extrapolation) of weighting factors for the step phantom. CNR values were calculated for different regions.

RESULTS

The CNR and SNR characteristics as functions of material cancellation and noise reduction weighting factors were obtained from theoretical models and experimental data from the step phantom. This showed a good qualitative validation of the models. For the ANCR algorithm, both the theory and experiment demonstrated that the material cancellation weighting factors (ω _Bn,ST ) can be optimized independently of the noise cancellation weighting factors (ω _A ), which can be optimized by maximizing SNR. For each ROI (with different overlapping bone and soft tissue thicknesses) the weighting factors ω _Bn,ST were determined as well as corresponding optimal weighting factors ω _A for noise reduction. For the Rando phantom, CNR values for regions representing different anatomical structures (ribs, spine, and tumor) were evaluated. It was shown that the proposed adaptive ACNR further improves image quality, compared to the conventional ACNR algorithm. The improvement is maximized for regions with bones (ribs or spine), where the largest attenuation is observed.

CONCLUSION

The ACNR weighting factors are dependent on the material thicknesses due to varying beam attenuation leading to different levels of quantum noise. This was shown with the derived theoretical expressions of the CNR and SNR functions and was validated by experimental data. The adaptive ANCR DE algorithm was developed, which allows for an increase in image quality by spatially varying weighting factors for noise reduction. This algorithm complements the previously developed PP-DE algorithm to obtain better quality DE images.

Characterization and potential applications of a dual-layer flat-panel detector

PURPOSE

Dual-energy (DE) x-ray imaging has many clinical applications in radiography, fluoroscopy, and CT. This work characterizes a prototype dual-layer (DL) flat-panel detector (FPD) and investigates its DE imaging capabilities for applications in two-dimensional (2D) radiography/fluoroscopy and quantitative three-dimensional (3D) cone-beam CT. Unlike other DE methods like kV switching, a DL FPD obtains DE images from a single exposure, making it robust against patient and system motion.

METHODS

The DL FPD consists of a top layer with a 200 µm-thick CsI scintillator coupled to an amorphous silicon (aSi) FPD of 150 µm pixel size and a bottom layer with a 550 µm thick CsI scintillator coupled to an identical aSi FPD. The two layers are separated by a 1-mm Cu filter to increase spectral separation. Images (43 × 43 cm2 active area) can be readout in 2 × 2 binning mode (300 µm pixels) at up to 15 frames per second. Detector performance was first characterized by measuring the MTF, NPS, and DQE for the top and bottom layers. For 2D applications, a qualitative study was conducted using an anthropomorphic thorax phantom containing a porcine heart with barium-filled coronary arteries (similar to iodine). Additionally, fluoroscopic lung tumor tracking was investigated by superimposing a moving tumor phantom on the thorax phantom. Tracking accuracies of single-energy (SE) and DE fluoroscopy were compared against the ground truth motion of the tumor. For 3D quantitative imaging, a phantom containing water, iodine, and calcium inserts was used to evaluate overall DE material decomposition capabilities. Virtual monoenergetic (VM) images ranging from 40 to 100 keV were generated, and the optimal VM image energy which achieved the highest image uniformity and maximum contrast-to-noise ratio (CNR) was determined.

RESULTS

The spatial resolution of the top layer was substantially higher than that of the bottom layer (top layer 50% MTF = 2.2 mm-1 , bottom layer = 1.2 mm-1 ). A substantial increase in NNPS and reduction in DQE were observed for the bottom layer mainly due to photon loss within the top layer and Cu filter. For 2D radiographic and fluoroscopic applications, the DL FPD was capable of generating high-quality material-specific images separating soft tissue from bone and barium. For lung tumor tracking, DE fluoroscopy yielded more accurate results than SE fluoroscopy, with an average reduction in the root mean square error (RMSE) of over 10×. For the DE-CBCT studies, accurate basis material decompositions were obtained. The estimated material densities were 294.68  ±  17.41 and 92.14  ±  15.61 mg/ml for the 300 and 100 mg/ml calcium inserts, respectively, and 8.93  ±  1.45, 4.72  ±  1.44, and 2.11  ±  1.32 mg/ml for the 10, 5, and 2 mg/ml iodine inserts, respectively, with an average error of less than 5%. The optimal VM image energy was found to be 60 keV.

CONCLUSIONS

We characterized a prototype DL FPD and demonstrated its ability to perform accurate single-exposure DE radiography/fluoroscopy and DE-CBCT. The merits of the DL detector approach include superior spatial and temporal registration between its constituent images, and less complicated acquisition sequences.

Adaptive weighted log subtraction based on neural networks for markerless tumor tracking using dual-energy fluoroscopy

PURPOSE

To present a novel method, based on convolutional neural networks (CNN), to automate weighted log subtraction (WLS) for dual-energy (DE) fluoroscopy to be used in conjunction with markerless tumor tracking (MTT).

METHODS

A CNN was developed to automate WLS (aWLS) of DE fluoroscopy to enhance soft tissue visibility. Briefly, this algorithm consists of two phases: training a CNN architecture to predict pixel-wise weighting factors followed by application of WLS subtraction to reduce anatomical noise. To train the CNN, a custom phantom was built consisting of aluminum (Al) and acrylic (PMMA) step wedges. Per-pixel ground truth (GT) weighting factors were calculated by minimizing the contrast of Al in the step wedge phantom to train the CNN. The pretrained model was then utilized to predict pixel-wise weighting factors for use in WLS. For comparison, the weighting factor was manually determined in each projection (mWLS). A thorax phantom with five simulated spherical targets (5-25 mm) embedded in a lung cavity, was utilized to assess aWLS performance. The phantom was imaged with fast-kV dual-energy (120 and 60 kVp) fluoroscopy using the on-board imager of a commercial linear accelerator. DE images were processed offline to produce soft tissue images using both WLS methods. MTT was compared using soft tissue images produced with both mWLS and aWLS techniques.

RESULTS

Qualitative evaluation demonstrated that both methods achieved soft tissue images with similar quality. The use of aWLS increased the number of tracked frames by 1-5% compared to mWLS, with the largest increase observed for the smallest simulated tumors. The tracking errors for both methods produced agreement to within 0.1 mm.

CONCLUSIONS

A novel method to perform automated WLS for DE fluoroscopy was developed. Having similar soft tissue quality as well as bone suppression capability as mWLS, this method allows for real-time processing of DE images for MTT.

Patient-specific pixel-based weighting factor dual-energy x-ray imaging system using a priori CT data

PURPOSE

The purpose of this study was to develop a novel patient-specific pixel-based weighting factor dual-energy (PP-DE) algorithm to effectively suppress bone throughout the image and overcome the limitation of the conventional DE algorithm with constant weighting factor which is restricted to regions with uniform patient thickness. Additionally, to derive theoretical expressions to describe the dependence of the weighting factors on several imaging parameters and validate them with measurement.

METHODS

A step phantom was constructed consisting of slabs of solid water and bone materials. Thicknesses of bone ranged [0-6] cm in one direction and solid water [5-30] cm in the other direction. Projection images at 60 and 140 kVp were acquired using a clinical imaging system. Optimal weighting factors were found by iteratively varying it in the range [0-1.4], where bone and soft-tissue contrast-to-noise ratio (CNR) reached zero. Bone and soft-tissue digitally reconstructed thicknesses were created using computed tomography (CT) images of a Rando phantom and ray tracing techniques. A weighting factor image (ω) was calculated using digitally reconstructed thicknesses (DRTs) and precalculated weighting factors from the step phantom. This ω image was then used to generate a PP-DE image. The PP-DE image was compared to the conventional DE image which uses a constant weighting factor throughout the image. The effect of the misaligned ω image on PP-DE images was investigated by acquiring LE and HE images at various shifts of Rando phantom. A rigid registration was used based on mutual information algorithm in Matlab. The signal-to-noise ratios (SNR) were calculated in the step phantom for the PP-DE image and compared to that of conventional DE technique. Analytical expressions for theoretical weighting factors were derived which included various effects such as beam hardening, scatter, and detector response. The analytical expressions were simulated in Spektr3.0 for different bone and solid water thicknesses as per the step phantom. A tray of steel pins was constructed and used with the step phantom to remove the scattered radiation. The simulated theoretical weighting factors were validated by comparing to those from the step phantom measurement.

RESULTS

Optimal weighting factor values for the step phantom varied from 0.633 to 1.372 depending on region thickness. Thicker regions required larger weighting factors for bone cancellation. The PP-DE image of the Rando phantom favorably cancelled both ribs and spine, whereas in the conventional DE image, only one could be cancelled at a time. The misaligned ω image was less effective in removing all bones indicating the importance of alignment as part of the PP-DE algorithm implementation. The SNRs for the PP-DE image was larger than those of the conventional DE images for regions which required smaller weighting factors for bone suppression. Comparisons of measured and simulated weighting factors demonstrated a 3% agreement for all bone overlapped regions except for the thickest region with 30 cm of solid water overlapped with 6 cm bone where the signal was lost due to excess attenuation.

CONCLUSIONS

A novel PP-DE algorithm was developed which can create higher quality DE images with enhanced bone cancellation and improved noise characteristics compared to conventional DE technique. In addition, theoretical weighting factor expressions were derived and validated against measurement.

Real-Time Whole-Brain Radiation Therapy: A Single-Institution Experience

PURPOSE

To demonstrate the feasibility of a real-time whole-brain radiation therapy (WBRT) workflow, taking advantage of contemporary radiation therapy capabilities and seeking to optimize clinical workflow for WBRT.

METHODS AND MATERIALS

We developed a method incorporating the linear accelerator's on-board imaging system for patient simulation, used cone-beam computed tomography (CBCT) data for treatment planning, and delivered the first fraction of prescribed therapy, all during the patient's initial appointment. Simulation was performed in the linear accelerator vault. An acquired CBCT data set was used for scripted treatment planning protocol, providing inversely planned, automated treatment plan generation. The osseous boundaries of the brain were auto-contoured to create a target volume. Two parallel-opposed beams using field-in-field intensity modulate radiation therapy covered this target to the user-defined inferior level (C1 or C2). The method was commissioned using an anthropomorphic head phantom and verified using 100 clinically treated patients.

RESULTS

Whole-brain target heterogeneity was within 95%-107% of the prescription dose, and target coverage compared favorably to standard, manually created 3-dimensional plans. For the commissioning CBCT datasets, the secondary monitor unit verification and independent 3-dimensional dose distribution comparison for computed and delivered doses were within 2% agreement relative to the scripted auto-plans. On average, time needed to complete the entire process was 35.1 ± 10.3 minutes from CBCT start to last beam delivered.

CONCLUSIONS

The real-time WBRT workflow using integrated on-site imaging, planning, quality assurance, and delivery was tested and deemed clinically feasible. The design necessitates a synchronized team consisting of physician, physicist, dosimetrist, and therapists. This work serves as a proof of concept of real-time planning and delivery for other treatment sites.

MRI-guided lung SBRT: Present and future developments

Stereotactic body radiotherapy (SBRT) is rapidly becoming an alternative to surgery for the treatment of early-stage non-small cell lung cancer patients. Lung SBRT is administered in a hypo-fractionated, conformal manner, delivering high doses to the target. To avoid normal-tissue toxicity, it is crucial to limit the exposure of nearby healthy organs-at-risk (OAR). Current image-guided radiotherapy strategies for lung SBRT are mostly based on X-ray imaging modalities. Although still in its infancy, magnetic resonance imaging (MRI) guidance for lung SBRT is not exposure-limited and MRI promises to improve crucial soft-tissue contrast. Looking beyond anatomical imaging, functional MRI is expected to inform treatment decisions and adaptations in the future. This review summarises and discusses how MRI could be advantageous to the different links of the radiotherapy treatment chain for lung SBRT: diagnosis and staging, tumour and OAR delineation, treatment planning, and inter- or intrafractional motion management. Special emphasis is placed on a new generation of hybrid MRI treatment devices and their potential for real-time adaptive radiotherapy.

Tumour auto-contouring on 2d cine MRI for locally advanced lung cancer: A comparative study

BACKGROUND AND PURPOSE

Radiotherapy guidance based on magnetic resonance imaging (MRI) is currently becoming a clinical reality. Fast 2d cine MRI sequences are expected to increase the precision of radiation delivery by facilitating tumour delineation during treatment. This study compares four auto-contouring algorithms for the task of delineating the primary tumour in six locally advanced (LA) lung cancer patients.

MATERIAL AND METHODS

Twenty-two cine MRI sequences were acquired using either a balanced steady-state free precession or a spoiled gradient echo imaging technique. Contours derived by the auto-contouring algorithms were compared against manual reference contours. A selection of eight image data sets was also used to assess the inter-observer delineation uncertainty.

RESULTS

Algorithmically derived contours agreed well with the manual reference contours (median Dice similarity index: ⩾0.91). Multi-template matching and deformable image registration performed significantly better than feature-driven registration and the pulse-coupled neural network (PCNN). Neither MRI sequence nor image orientation was a conclusive predictor for algorithmic performance. Motion significantly degraded the performance of the PCNN. The inter-observer variability was of the same order of magnitude as the algorithmic performance.

CONCLUSION

Auto-contouring of tumours on cine MRI is feasible in LA lung cancer patients. Despite large variations in implementation complexity, the different algorithms all have relatively similar performance.

A Bayesian approach for three-dimensional markerless tumor tracking using kV imaging during lung radiotherapy

The ability to monitor tumor motion without implanted markers can potentially enable broad access to more accurate and precise lung radiotherapy. A major challenge is that kilovoltage (kV) imaging based methods are rarely able to continuously track the tumor due to the inferior tumor visibility on 2D kV images. Another challenge is the estimation of 3D tumor position based on only 2D imaging information. The aim of this work is to address both challenges by proposing a Bayesian approach for markerless tumor tracking for the first time. The proposed approach adopts the framework of the extended Kalman filter, which combines a prediction and measurement steps to make the optimal tumor position update. For each imaging frame, the tumor position is first predicted by a respiratory-correlated model. The 2D tumor position on the kV image is then measured by template matching. Finally, the prediction and 2D measurement are combined based on the 3D distribution of tumor positions in the past 10 s and the estimated uncertainty of template matching. To investigate the clinical feasibility of the proposed method, a total of 13 lung cancer patient datasets were used for retrospective validation, including 11 cone-beam CT scan pairs and two stereotactic ablative body radiotherapy cases. The ground truths for tumor motion were generated from the the 3D trajectories of implanted markers or beacons. The mean, standard deviation, and 95th percentile of the 3D tracking error were found to range from 1.6-2.9 mm, 0.6-1.5 mm, and 2.6-5.8 mm, respectively. Markerless tumor tracking always resulted in smaller errors compared to the standard of care. The improvement was the most pronounced in the superior-inferior (SI) direction, with up to 9.5 mm reduction in the 95th-percentile SI error for patients with  >10 mm 5th-to-95th percentile SI tumor motion. The percentage of errors with 3D magnitude  <5 mm was 96.5% for markerless tumor tracking and 84.1% for the standard of care. The feasibility of 3D markerless tumor tracking has been demonstrated on realistic clinical scenarios for the first time. The clinical implementation of the proposed method will enable more accurate and precise lung radiotherapy using existing hardware and workflow. Future work is focused on the clinical and real-time implementation of this method.

Evaluation of a template-based algorithm for markerless lung tumour localization on single- and dual-energy kilovoltage images

OBJECTIVE

To evaluate a template-based matching algorithm on single-energy (SE) and dual-energy (DE) radiographs for markerless localization of lung tumours.

METHODS

A total of 74 images from 17 patients with Stages IA-IV lung cancer were considered. At the time of radiotherapy treatment, gated end-expiration SE radiographs were obtained at 60 and 120 kVp at different gantry angles (33° anterior and 41° oblique), from which soft-tissue-enhanced DE images were created. A template-based matching algorithm was used to localize individual tumours on both SE and DE radiographs. Tumour centroid co-ordinates obtained from the template-matching software on both SE and DE images were compared with co-ordinates defined by physicians.

RESULTS

The template-based matching algorithm was able to successfully localize the gross tumor volume within 5 mm on 70% (52/74) of the SE images vs 91% (66/74) of the DE images (p < 0.01). The mean vector differences between the co-ordinates of the template matched by the algorithm and the co-ordinates of the physician-defined ground truth were 3.2 ± 2.8 mm for SE images vs 2.3 ± 1.7 mm for DE images (p = 0.03).

CONCLUSION

Template-based matching on DE images was more accurate and precise than using SE images. Advances in knowledge: This represents, to the authors' knowledge, the largest study evaluating template matching on clinical SE and DE images, considering not only anterior gantry angles but also oblique angles, suggesting a novel lung tumour matching technique using DE subtraction that is reliable, accurate and precise.

Using dual-energy x-ray imaging to enhance automated lung tumor tracking during real-time adaptive radiotherapy

PURPOSE

Real-time, markerless localization of lung tumors with kV imaging is often inhibited by ribs obscuring the tumor and poor soft-tissue contrast. This study investigates the use of dual-energy imaging, which can generate radiographs with reduced bone visibility, to enhance automated lung tumor tracking for real-time adaptive radiotherapy.

METHODS

kV images of an anthropomorphic breathing chest phantom were experimentally acquired and radiographs of actual lung cancer patients were Monte-Carlo-simulated at three imaging settings: low-energy (70 kVp, 1.5 mAs), high-energy (140 kVp, 2.5 mAs, 1 mm additional tin filtration), and clinical (120 kVp, 0.25 mAs). Regular dual-energy images were calculated by weighted logarithmic subtraction of high- and low-energy images and filter-free dual-energy images were generated from clinical and low-energy radiographs. The weighting factor to calculate the dual-energy images was determined by means of a novel objective score. The usefulness of dual-energy imaging for real-time tracking with an automated template matching algorithm was investigated.

RESULTS

Regular dual-energy imaging was able to increase tracking accuracy in left-right images of the anthropomorphic phantom as well as in 7 out of 24 investigated patient cases. Tracking accuracy remained comparable in three cases and decreased in five cases. Filter-free dual-energy imaging was only able to increase accuracy in 2 out of 24 cases. In four cases no change in accuracy was observed and tracking accuracy worsened in nine cases. In 9 out of 24 cases, it was not possible to define a tracking template due to poor soft-tissue contrast regardless of input images. The mean localization errors using clinical, regular dual-energy, and filter-free dual-energy radiographs were 3.85, 3.32, and 5.24 mm, respectively. Tracking success was dependent on tumor position, tumor size, imaging beam angle, and patient size.

CONCLUSIONS

This study has highlighted the influence of patient anatomy on the success rate of real-time markerless tumor tracking using dual-energy imaging. Additionally, the importance of the spectral separation of the imaging beams used to generate the dual-energy images has been shown.