Deep learning-based 4D-synthetic CTs from sparse-view CBCTs for dose calculations in adaptive proton therapy

Quantitative use of cone-beam computed tomography in proton therapy: challenges and opportunities

The fundamental goal in radiation therapy (RT) is to simultaneously maximize tumor cell killing and healthy tissue sparing. Reducing uncertainty margins improves normal tissue sparing, but generally requires advanced techniques. Adaptive RT (ART) is a compelling technique that leverages daily imaging and anatomical information to support reduced margins and to optimize plan quality for each treatment fraction. An especially exciting avenue for ART is proton therapy (PT), which aims to combine daily plan re-optimization with the unique advantages provided by protons, including reduced integral dose and near-zero dose deposition distal to the target along the beam direction. A core component for ART is onboard image guidance, and currently two options are available on proton systems, including cone-beam computed tomography (CBCT) and CT-on-rail (CToR) imaging. While CBCT suffers from poorer image quality compared to CToR imaging, CBCT platforms can be more easily integrated with PT systems and thus may support more streamlined adaptive proton therapy (APT). In this review, we present current status of CBCT application to proton therapy dose evaluation and plan adaptation, including progress, challenges and future directions.

MBST-Driven 4D-CBCT reconstruction: Leveraging swin transformer and masking for robust performance

OBJECTIVE

This research developed an innovative Mask-based Swin Transformer network (MBST) to enhance the quality of 4D cone-beam computed tomography (4D-CBCT) reconstruction. The network is trained on 4D-CBCT reconstructed under limited scanning conditions, enabling its application to a broad range of 4D-CBCT reconstruction scenarios, including those with high scanning speeds.

METHODS

4D imaging data from 20 patients with thoracic tumors were used to train and evaluate the deep learning model. 15 cases were used for training, and 5 cases were employed for simulation testing. The Feldkamp-Davis-Kress algorithm was employed to simulate 4D-CBCT from downsampled 4D-CT data to mitigate the uncertainties associated with respiratory motion between treatment fractions, and the 4D-CT data served as the ground truth for training. The study reconstructed 4D-CBCT images under 11 different scanning intervals including full angle acquisition at 1°, 2°, 3°, 4°, 5°, 6°, 12°, 18°, 24° intervals, and 1/3 full angles acquisition at 5°, 10° inrevals respectively for capturing 4D-CBCT projections. The test results were quantitatively evaluated using the structural similarity index measure (SSIM), peak signal-to-noise ratio (PSNR), mean error (ME), and mean absolute error (MAE), and image quality was qualitatively assessed. Real clinical patients who were not included in the training were tested to evaluate the network's ability to generalize. Moreover, the proposed method was compared with other deep learning approaches, and statistical analyses were performed.

RESULTS

Simulation data assessment revealed that with small projection acquisition interval, such as the 4°interval, the 4D-CBCT images optimized by MBST showed a considerable improvement over the original 4D-CBCT images in terms of SSIM (42.3% increase) and PSNR (10.8 dB increase), and the ME and MAE values approached 0. The improvements were statistically significant (P < 0.001). Compared with other deep learning methods, MBST demonstrated superior performance with improvements of 1.4% in SSIM and 1.21 dB in PSNR and a reduction of 0.94 in MAE. With large projection intervals, such as the 24°interval, MBST outperformed other deep learning methods. Specifically, its SSIM, PSNR, and MAE increased by 3.8%, 0.81 dB, and 10.34, respectively, compared with those of other deep learning methods, and the improvements were statistically significant (P < 0.01). In addition, MBST could reconstruct bone tissue and optimize the quality of 4D-CBCT images even when the number of projections was small (12°, 18°, 24°intervals). Clinical data evaluation revealed that after optimization by MBST, the SSIM, PSNR, ME, and MAE of 4D-CBCT compared with those of 4D-CT registration improved from the original 22.8%, 15.49 dB, -345.5, and 432.2 to 81.5%, 27.93 dB, -53.79, and 73.77, respectively. Moreover, MBST exhibited the most pronounced improvement among all the compared methods. MBST could accurately recover high-density structure, lung structures, and tracheal walls.

CONCLUSION

This study comprehensively demonstrated the ability of MBST to reconstruct 4D-CBCT images under various scanning conditions. When the method was tested on clinical patient datasets, its CT values and image quality achieved satisfactory results. Thus, MBST can serve as a highly generalized reconstruction network for improving the quality of 4D-CBCT images.

Artificial intelligence for treatment delivery: image-guided radiotherapy

Radiation therapy (RT) is a highly digitized field relying heavily on computational methods and, as such, has a high affinity for the automation potential afforded by modern artificial intelligence (AI). This is particularly relevant where imaging is concerned and is especially so during image-guided RT (IGRT). With the advent of online adaptive RT (ART) workflows at magnetic resonance (MR) linear accelerators (linacs) and at cone-beam computed tomography (CBCT) linacs, the need for automation is further increased. AI as applied to modern IGRT is thus one area of RT where we can expect important developments in the near future. In this review article, after outlining modern IGRT and online ART workflows, we cover the role of AI in CBCT and MRI correction for dose calculation, auto-segmentation on IGRT imaging, motion management, and response assessment based on in-room imaging.

Proton therapy of lung cancer patients - Treatment strategies and clinical experience from a medical physicist's perspective

PURPOSE

Proton therapy of moving targets is considered a challenge. At Maastro, we started treating lung cancer patients with proton therapy in October 2019. In this work, we summarise the developed treatment strategies and gained clinical experience from a physics point of view.

METHODS

We report on our clinical approaches to treat lung cancer patients with the Mevion Hyperscan S250i proton machine. We classify lung cancer patients as small movers (tumour movement ≤ 5 mm) or large movers (tumour movement > 5 mm). The preferred beam configuration has evolved over the years of clinical treatment, and currently mostly two or three beam directions are used. All patients are treated with robustly optimised plans (5 mm setup and 3% range uncertainty). Small movers are planned based on a clinical target volume (CTV) with a 3 mm isotropic margin expansion to account for motion, while large movers are planned based on an internal target volume (ITV). All patients are treated in free-breathing.

RESULTS

Between October 2019 and December 2023, 379 lung cancer patients have been treated, of which 130 were large movers. The adaptation rate was 28%. The median treatment time has been reduced from 30 to 23 min. The mean dose to the heart, oesophagus, and lungs was on average 4.3, 15.4, and 11.0 Gy, respectively.

CONCLUSIONS

Several treatment planning and workflow improvements have been introduced over the years, resulting in an increase of treatment quality and number of treated patients, as well as reduction of planning and treatment time.

The emerging role of Artificial Intelligence in proton therapy: A review

Artificial intelligence (AI) has made a tremendous impact in the space of healthcare, and proton therapy is not an exception. Proton therapy has witnessed growing popularity in oncology over recent decades, and researchers are increasingly looking to develop AI and machine learning tools to aid in various steps of the treatment planning and delivery processes. This review delves into the emergent role of AI in proton therapy, evaluating its development, advantages, intended clinical contexts, and areas of application. Through the analysis of 76 studies, we aim to underscore the importance of AI applications in advancing proton therapy and to highlight their prospective influence on clinical practices.

Adaptive intensity modulated proton therapy using 4D robust planning: a proof-of-concept for the application of dose mimicking approach

Objective.A four-dimensional robust optimisation (4DRO) is usually employed when the tumour respiratory motion needs to be addressed. However, it is computationally demanding, and an automated method is preferable for adaptive planning to avoid manual trial-and-error. This study proposes a 4DRO technique based on dose mimicking for automated adaptive planning.Approach.Initial plans for 4DRO intensity modulated proton therapy were created on an average CT for four patients with clinical target volume (CTV) in the lung, oesophagus, or pancreas, respectively. These plans were robustly optimised using three phases of four-dimensional computed tomography (4DCT) and accounting for setup and density uncertainties. Weekly 4DCTs were used for adaptive replanning, using a constant relative biological effectiveness (cRBE) of 1.1. Two methods were used: (1) template-based adaptive (TA) planning and (2) dose-mimicking-based adaptive (MA) planning. The plans were evaluated using variable RBE (vRBE) weighted doses and biologically consistent dose accumulation (BCDA).Main results.MA and TA plans had comparable CTV coverage except for one patient where the MA plan had a higher D98 and lower D2 but with an increased D2 in few organs at risk (OARs). CTV D98 deviations in non-adaptive plans from the initial plans were up to -7.2 percentage points (p.p.) in individual cases and -1.8 p.p. when using BCDA. For the OARs, MA plans showed a reduced mean dose and D2 compared to the TA plans, with few exceptions. The vRBE-weighted accumulated doses had a mean dose and D2 difference of up to 0.3 Gy and 0.5 Gy, respectively, in the OARs with respect to cRBE-weighted doses.Significance.MA plans indicate better performance in target coverage and OAR dose sparing compared to the TA plans in 4DRO adaptive planning. Moreover, MA method is capable of handling both forms of anatomical variation, namely, changes in density and relative shifts in the position of OARs.

A 4D-CBCT correction network based on contrastive learning for dose calculation in lung cancer

OBJECTIVE

This study aimed to present a deep-learning network called contrastive learning-based cycle generative adversarial networks (CLCGAN) to mitigate streak artifacts and correct the CT value in four-dimensional cone beam computed tomography (4D-CBCT) for dose calculation in lung cancer patients.

METHODS

4D-CBCT and 4D computed tomography (CT) of 20 patients with locally advanced non-small cell lung cancer were used to paired train the deep-learning model. The lung tumors were located in the right upper lobe, right lower lobe, left upper lobe, and left lower lobe, or in the mediastinum. Additionally, five patients to create 4D synthetic computed tomography (sCT) for test. Using the 4D-CT as the ground truth, the quality of the 4D-sCT images was evaluated by quantitative and qualitative assessment methods. The correction of CT values was evaluated holistically and locally. To further validate the accuracy of the dose calculations, we compared the dose distributions and calculations of 4D-CBCT and 4D-sCT with those of 4D-CT.

RESULTS

The structural similarity index measure (SSIM) and peak signal-to-noise ratio (PSNR) of the 4D-sCT increased from 87% and 22.31 dB to 98% and 29.15 dB, respectively. Compared with cycle consistent generative adversarial networks, CLCGAN enhanced SSIM and PSNR by 1.1% (p < 0.01) and 0.42% (p < 0.01). Furthermore, CLCGAN significantly decreased the absolute mean differences of CT value in lungs, bones, and soft tissues. The dose calculation results revealed a significant improvement in 4D-sCT compared to 4D-CBCT. CLCGAN was the most accurate in dose calculations for left lung (V5Gy), right lung (V5Gy), right lung (V20Gy), PTV (D98%), and spinal cord (D2%), with the relative dose difference were reduced by 6.84%, 3.84%, 1.46%, 0.86%, 3.32% compared to 4D-CBCT.

CONCLUSIONS

Based on the satisfactory results obtained in terms of image quality, CT value measurement, it can be concluded that CLCGAN-based corrected 4D-CBCT can be utilized for dose calculation in lung cancer.

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.

Technical note: Evaluation of deep learning based synthetic CTs clinical readiness for dose and NTCP driven head and neck adaptive proton therapy

BACKGROUND

Adaptive proton therapy workflows rely on accurate imaging throughout the treatment course. Our centre currently utilizes weekly repeat CTs (rCTs) for treatment monitoring and plan adaptations. However, deep learning-based methods have recently shown to successfully correct CBCT images, which suffer from severe imaging artifacts, and generate high quality synthetic CT (sCT) images which enable CBCT-based proton dose calculations.

PURPOSE

To compare daily CBCT-based sCT images to planning CTs (pCT) and rCTs of head and neck (HN) cancer patients to investigate the dosimetric accuracy of CBCT-based sCTs in a scenario mimicking actual clinical practice.

METHODS

Data of 56 HN cancer patients, previously treated with proton therapy was used to generate 1.962 sCT images, using a previously developed and trained deep convolutional neural network. Clinical IMPT treatment plans were recalculated on the pCT, weekly rCTs and daily sCTs. The dosimetric accuracy of sCTs was compared to same day rCTs and the initial planning CT. As a reference, rCTs were also compared to pCTs. The dose difference between sCTs and rCTs/pCT was quantified by calculating the D98 difference for target volumes and Dmean difference for organs-at-risk. To investigate the clinical relevancy of possible dose differences, NTCP values were calculated for dysphagia and xerostomia.

RESULTS

For target volumes, only minor dose differences were found for sCT versus rCT and sCT versus pCT, with dose differences mostly within ±1.5%. Larger dose differences were observed in OARs, where a general shift towards positive differences was found, with the largest difference in the left parotid gland. Delta NTCP values for grade 2 dysphagia and xerostomia were within ±2.5% for 90% of the sCTs.

CONCLUSIONS

Target doses showed high similarity between rCTs and sCTs. Further investigations are required to identify the origin of the dose differences at OAR levels and its relevance in clinical decision making.

Comparison and evaluation of different deep learning models of synthetic CT generation from CBCT for nasopharynx cancer adaptive proton therapy

BACKGROUND

Cone-beam computed tomography (CBCT) scanning is used for patient setup in image-guided radiotherapy. However, its inaccurate CT numbers limit its applicability in dose calculation and treatment planning.

PURPOSE

This study compares four deep learning methods for generating synthetic CT (sCT) to determine which method is more appropriate and offers potential for further clinical exploration in adaptive proton therapy for nasopharynx cancer.

METHODS

CBCTs and deformed planning CT (dCT) from 75 patients (60/5/10 for training, validation and testing) were used to compare cycle-consistent Generative Adversarial Network (cycleGAN), Unet, Unet+cycleGAN and conditionalGenerative Adversarial Network (cGAN) for sCT generation. The sCT images generated by each method were evaluated against dCT images using mean absolute error (MAE), structural similarity (SSIM), peak signal-to-noise ratio (PSNR), spatial non-uniformity (SNU) and radial averaging in the frequency domain. In addition, dosimetric accuracy was assessed through gamma analysis, differences in water equivalent thickness (WET), and dose-volume histogram metrics.

RESULTS

The cGAN model has demonstrated optimal performance in the four models across various indicators. In terms of image quality under global condition, the average MAE has been reduced to 16.39HU, SSIM has increased to 95.24%, and PSNR has increased to 28.98. Regarding dosimetric accuracy, the gamma passing rate (2%/2 mm) has reached 99.02%, and the WET difference is only 1.28 mm. The D95 value of CTVs coverage and Dmax value of spinal cord, brainstem show no significant differences between dCT and sCT generated by cGAN model.

CONCLUSIONS

The cGAN model has been shown to be a more suitable approach for generating sCT using CBCT, considering its characteristics and concepts. The resulting sCT has the potential for application in adaptive proton therapy.

Clinical evaluation of synthetic computed tomography methods in adaptive proton therapy of lung cancer patients

Background and purpose

Efficient workflows for adaptive proton therapy are of high importance. This study evaluated the possibility to replace repeat-CTs (reCTs) with synthetic CTs (sCTs), created based on cone-beam CTs (CBCTs), for flagging the need of plan adaptations in intensity-modulated proton therapy (IMPT) treatment of lung cancer patients.

Materials and methods

Forty-two IMPT patients were retrospectively included. For each patient, one CBCT and a same-day reCT were included. Two commercial sCT methods were applied; one based on CBCT number correction (Cor-sCT), and one based on deformable image registration (DIR-sCT). The clinical reCT workflow (deformable contour propagation and robust dose re-computation) was performed on the reCT as well as the two sCTs. The deformed target contours on the reCT/sCTs were checked by radiation oncologists and edited if needed. A dose-volume-histogram triggered plan adaptation method was compared between the reCT and the sCTs; patients needing a plan adaptation on the reCT but not on the sCT were denoted false negatives. As secondary evaluation, dose-volume-histogram comparison and gamma analysis (2%/2mm) were performed between the reCT and sCTs.

Results

There were five false negatives, two for Cor-sCT and three for DIR-sCT. However, three of these were only minor, and one was caused by tumour position differences between the reCT and CBCT and not by sCT quality issues. An average gamma pass rate of 93% was obtained for both sCT methods.

Conclusion

Both sCT methods were judged to be of clinical quality and valuable for reducing the amount of reCT acquisitions.

Emerging technologies for cancer therapy using accelerated particles

Cancer therapy with accelerated charged particles is one of the most valuable biomedical applications of nuclear physics. The technology has vastly evolved in the past 50 years, the number of clinical centers is exponentially growing, and recent clinical results support the physics and radiobiology rationale that particles should be less toxic and more effective than conventional X-rays for many cancer patients. Charged particles are also the most mature technology for clinical translation of ultra-high dose rate (FLASH) radiotherapy. However, the fraction of patients treated with accelerated particles is still very small and the therapy is only applied to a few solid cancer indications. The growth of particle therapy strongly depends on technological innovations aiming to make the therapy cheaper, more conformal and faster. The most promising solutions to reach these goals are superconductive magnets to build compact accelerators; gantryless beam delivery; online image-guidance and adaptive therapy with the support of machine learning algorithms; and high-intensity accelerators coupled to online imaging. Large international collaborations are needed to hasten the clinical translation of the research results.

ScatterNet for projection-based 4D cone-beam computed tomography intensity correction of lung cancer patients

Background and purpose: In radiotherapy, dose calculations based on 4D cone beam CTs (4DCBCTs) require image intensity corrections. This retrospective study compared the dose calculation accuracy of a deep learning, projection-based scatter correction workflow (ScatterNet), to slower workflows: conventional 4D projection-based scatter correction (CBCTcor) and a deformable image registration (DIR)-based method (4DvCT). Materials and methods: For 26 lung cancer patients, planning CTs (pCTs), 4DCTs and CBCT projections were available. ScatterNet was trained with pairs of raw and corrected CBCT projections. Corrected projections from ScatterNet and the conventional workflow were reconstructed using MA-ROOSTER, yielding 4DCBCTSN and 4DCBCTcor. The 4DvCT was generated by 4DCT to 4DCBCT DIR, as part of the 4DCBCTcor workflow. Robust intensity modulated proton therapy treatment plans were created on free-breathing pCTs. 4DCBCTSN was compared to 4DCBCTcor and the 4DvCT in terms of image quality and dose calculation accuracy (dose-volume-histogram parameters and 3 % /3 mm gamma analysis). Results: 4DCBCTSN resulted in an average mean absolute error of 87 HU and 102 HU when compared to 4DCBCTcor and 4DvCT respectively. High agreement was observed in targets with median dose differences of 0.4 Gy (4DCBCTSN-4DCBCTcor) and 0.3 Gy (4DCBCTSN-4DvCT). The gamma analysis showed high average 3 % /3 mm pass rates of 96 % for both 4DCBCTSN vs. 4DCBCTcor and 4DCBCTSN vs. 4DvCT. Conclusions: Accurate 4D dose calculations are feasible for lung cancer patients using ScatterNet for 4DCBCT correction. Average scatter correction times could be reduced from 10 min (4DCBCTcor) to 3.9 s , showing the clinical suitability of the proposed deep learning-based method.

Application of Artificial Intelligence in the Diagnosis, Treatment, and Prognostic Evaluation of Mediastinal Malignant Tumors

Artificial intelligence (AI), also known as machine intelligence, is widely utilized in the medical field, promoting medical advances. Malignant tumors are the critical focus of medical research and improvement of clinical diagnosis and treatment. Mediastinal malignancy is an important tumor that attracts increasing attention today due to the difficulties in treatment. Combined with artificial intelligence, challenges from drug discovery to survival improvement are constantly being overcome. This article reviews the progress of the use of AI in the diagnosis, treatment, and prognostic prospects of mediastinal malignant tumors based on current literature findings.

Inter- and intrafractional 4D dose accumulation for evaluating ΔNTCP robustness in lung cancer

BACKGROUND AND PURPOSE

Model-based selection of proton therapy patients relies on a predefined reduction in normal tissue complication probability (NTCP) with respect to photon therapy. The decision is necessarily made based on the treatment plan, but NTCP can be affected when the delivered treatment deviates from the plan due to delivery inaccuracies. Especially for proton therapy of lung cancer, this can be important because of tissue density changes and, with pencil beam scanning, the interplay effect between the proton beam and breathing motion.

MATERIALS AND METHODS

In this work, we verified whether the expected benefit of proton therapy is retained despite delivery inaccuracies by reconstructing the delivered treatment using log-file based dose reconstruction and inter- and intrafractional accumulation. Additionally, the importance of two uncertain parameters for treatment reconstruction, namely deformable image registration (DIR) algorithm and α/β ratio, was assessed.

RESULTS

The expected benefit or proton therapy was confirmed in 97% of all studied cases, despite regular differences up to 2 percent point (p.p.) NTCP between the delivered and planned treatments. The choice of DIR algorithm affected NTCP up to 1.6 p.p., an order of magnitude higher than the effect of α/β ratio.

CONCLUSION

For the patient population and treatment technique employed, the predicted clinical benefit for patients selected for proton therapy was confirmed for 97.0% percent of all cases, although the NTCP based proton selection was subject to 2 p.p. variations due to delivery inaccuracies.

Improving CBCT image quality to the CT level using RegGAN in esophageal cancer adaptive radiotherapy

OBJECTIVE

This study aimed to improve the image quality and CT Hounsfield unit accuracy of daily cone-beam computed tomography (CBCT) using registration generative adversarial networks (RegGAN) and apply synthetic CT (sCT) images to dose calculations in radiotherapy.

METHODS

The CBCT/planning CT images of 150 esophageal cancer patients undergoing radiotherapy were used for training (120 patients) and testing (30 patients). An unsupervised deep-learning method, the 2.5D RegGAN model with an adaptively trained registration network, was proposed, through which sCT images were generated. The quality of deep-learning-generated sCT images was quantitatively compared to the reference deformed CT (dCT) image using mean absolute error (MAE), root mean square error (RMSE) of Hounsfield units (HU), and peak signal-to-noise ratio (PSNR). The dose calculation accuracy was further evaluated for esophageal cancer radiotherapy plans, and the same plans were calculated on dCT, CBCT, and sCT images.

RESULTS

The quality of sCT images produced by RegGAN was significantly improved compared to the original CBCT images. ReGAN achieved image quality in the testing patients with MAE sCT vs. CBCT: 43.7 ± 4.8 vs. 80.1 ± 9.1; RMSE sCT vs. CBCT: 67.2 ± 12.4 vs. 124.2 ± 21.8; and PSNR sCT vs. CBCT: 27.9 ± 5.6 vs. 21.3 ± 4.2. The sCT images generated by the RegGAN model showed superior accuracy on dose calculation, with higher gamma passing rates (93.3 ± 4.4, 90.4 ± 5.2, and 84.3 ± 6.6) compared to original CBCT images (89.6 ± 5.7, 85.7 ± 6.9, and 72.5 ± 12.5) under the criteria of 3 mm/3%, 2 mm/2%, and 1 mm/1%, respectively.

CONCLUSION

The proposed deep-learning RegGAN model seems promising for generation of high-quality sCT images from stand-alone thoracic CBCT images in an efficient way and thus has the potential to support CBCT-based esophageal cancer adaptive radiotherapy.