Development of a chest X-ray machine learning convolutional neural network model on a budget and using artificial intelligence explainability techniques to analyze patterns of machine learning inference

Objective

Machine learning (ML) will have a large impact on medicine and accessibility is important. This study's model was used to explore various concepts including how varying features of a model impacted behavior.

Materials and Methods

This study built an ML model that classified chest X-rays as normal or abnormal by using ResNet50 as a base with transfer learning. A contrast enhancement mechanism was implemented to improve performance. After training with a dataset of publicly available chest radiographs, performance metrics were determined with a test set. The ResNet50 base was substituted with deeper architectures (ResNet101/152) and visualization methods used to help determine patterns of inference.

Results

Performance metrics were an accuracy of 79%, recall 69%, precision 96%, and area under the curve of 0.9023. Accuracy improved to 82% and recall to 74% with contrast enhancement. When visualization methods were applied and the ratio of pixels used for inference measured, deeper architectures resulted in the model using larger portions of the image for inference as compared to ResNet50.

Discussion

The model performed on par with many existing models despite consumer-grade hardware and smaller datasets. Individual models vary thus a single model's explainability may not be generalizable. Therefore, this study varied architecture and studied patterns of inference. With deeper ResNet architectures, the machine used larger portions of the image to make decisions.

Conclusion

An example using a custom model showed that AI (Artificial Intelligence) can be accessible on consumer-grade hardware, and it also demonstrated an example of studying themes of ML explainability by varying ResNet architectures.

Adaptive scatter kernel deconvolution modeling for cone-beam CT scatter correction via deep reinforcement learning

BACKGROUND

Scattering photons can seriously contaminate cone-beam CT (CBCT) image quality with severe artifacts and substantial degradation of CT value accuracy, which is a major concern limiting the widespread application of CBCT in the medical field. The scatter kernel deconvolution (SKD) method commonly used in clinic requires a Monte Carlo (MC) simulation to determine numerous quality-related kernel parameters, and it cannot realize intelligent scatter kernel parameter optimization, causing limited accuracy of scatter estimation.

PURPOSE

Aiming at improving the scatter estimation accuracy of the SKD algorithm, an intelligent scatter correction framework integrating the SKD with deep reinforcement learning (DRL) scheme is proposed.

METHODS

Our method firstly builds a scatter kernel model to iteratively convolve with raw projections, and then the deep Q-network of the DRL scheme is introduced to intelligently interact with the scatter kernel to achieve a projection adaptive parameter optimization. The potential of the proposed framework is demonstrated on CBCT head and pelvis simulation data and experimental CBCT measurement data. Furthermore, we have implemented the U-net based scatter estimation approach for comparison.

RESULTS

The simulation study demonstrates that the mean absolute percentage error (MAPE) of the proposed method is less than 9.72% and the peak signal-to-noise ratio (PSNR) is higher than 23.90 dB, while for the conventional SKD algorithm, the minimum MAPE is 17.92% and the maximum PSNR is 19.32 dB. In the measurement study, we adopt a hardware-based beam stop array algorithm to obtain the scatter-free projections as a comparison baseline, and our method can achieve superior performance with MAPE < 17.79% and PSNR > 16.34 dB.

CONCLUSIONS

In this paper, we propose an intelligent scatter correction framework that integrates the physical scatter kernel model with DRL algorithm, which has the potential to improve the accuracy of the clinical scatter correction method to obtain better CBCT imaging quality.

Cone beam CT-based adaptive intensity modulated proton therapy assessment using automated planning for head-and-neck cancer

BACKGROUND

To assess the feasibility of CBCT-based adaptive intensity modulated proton therapy (IMPT) using automated planning for treatment of head and neck (HN) cancers.

METHODS

Twenty HN cancer patients who received radiotherapy and had pretreatment CBCTs were included in this study. Initial IMPT plans were created using automated planning software for all patients. Synthetic CTs (sCT) were then created by deforming the planning CT (pCT) to the pretreatment CBCTs. To assess dose calculation accuracy on sCTs, repeat CTs (rCTs) were deformed to the pretreatment CBCT obtained on the same day to create deformed rCT (rCTdef), serving as gold standard. The dose recalculated on sCT and on rCTdef were compared by using Gamma analysis. The accuracy of DIR generated contours was also assessed. To explore the potential benefits of adaptive IMPT, two sets of plans were created for each patient, a non-adapted IMPT plan and an adapted IMPT plan calculated on weekly sCT images. The weekly doses for non-adaptive and adaptive IMPT plans were accumulated on the pCT, and the accumulated dosimetric parameters of two sets were compared.

RESULTS

Gamma analysis of the dose recalculated on sCT and rCTdef resulted in a passing rate of 97.9% ± 1.7% using 3 mm/3% criteria. With the physician-corrected contours on the sCT, the dose deviation range of using sCT to estimate mean dose for the most organ at risk (OARs) can be reduced to (- 2.37%, 2.19%) as compared to rCTdef, while for V95 of primary or secondary CTVs, the deviation can be controlled within (- 1.09%, 0.29%). Comparison of the accumulated doses from the adaptive planning against the non-adaptive plans reduced mean dose to constrictors (- 1.42 Gy ± 2.79 Gy) and larynx (- 2.58 Gy ± 3.09 Gy). The reductions result in statistically significant reductions in the normal tissue complication probability (NTCP) of larynx edema by 7.52% ± 13.59%. 4.5% of primary CTVs, 4.1% of secondary CTVs, and 26.8% tertiary CTVs didn't meet the V95 > 95% constraint on non-adapted IMPT plans. All adaptive plans were able to meet the coverage constraint.

CONCLUSION

sCTs can be a useful tool for accurate proton dose calculation. Adaptive IMPT resulted in better CTV coverage, OAR sparing and lower NTCP for some OARs as compared with non-adaptive IMPT.

MB-DECTNet: a model-based unrolling network for accurate 3D dual-energy CT reconstruction from clinically acquired helical scans

Objective.Over the past several decades, dual-energy CT (DECT) imaging has seen significant advancements due to its ability to distinguish between materials. DECT statistical iterative reconstruction (SIR) has exhibited potential for noise reduction and enhanced accuracy. However, its slow convergence and substantial computational demands render the elapsed time for 3D DECT SIR often clinically unacceptable. The objective of this study is to accelerate 3D DECT SIR while maintaining subpercentage or near-subpercentage accuracy.Approach.We incorporate DECT SIR into a deep-learning model-based unrolling network for 3D DECT reconstruction (MB-DECTNet), which can be trained end-to-end. This deep learning-based approach is designed to learn shortcuts between initial conditions and the stationary points of iterative algorithms while preserving the unbiased estimation property of model-based algorithms. MB-DECTNet comprises multiple stacked update blocks, each containing a data consistency layer (DC) and a spatial mixer layer, with the DC layer functioning as a one-step update from any traditional iterative algorithm.Main results.The quantitative results indicate that our proposed MB-DECTNet surpasses both the traditional image-domain technique (MB-DECTNet reduces average bias by a factor of 10) and a pure deep learning method (MB-DECTNet reduces average bias by a factor of 8.8), offering the potential for accurate attenuation coefficient estimation, akin to traditional statistical algorithms, but with considerably reduced computational costs. This approach achieves 0.13% bias and 1.92% mean absolute error and reconstructs a full image of a head in less than 12 min. Additionally, we show that the MB-DECTNet output can serve as an initializer for DECT SIR, leading to further improvements in results.Significance.This study presents a model-based deep unrolling network for accurate 3D DECT reconstruction, achieving subpercentage error in estimating virtual monoenergetic images for a full head at 60 and 150 keV in 30 min, representing a 40-fold speedup compared to traditional approaches. These findings have significant implications for accelerating DECT SIR and making it more clinically feasible.

An unsupervised dual contrastive learning framework for scatter correction in cone-beam CT image

PURPOSE

Cone-beam computed tomography (CBCT) is widely utilized in modern radiotherapy; however, CBCT images exhibit increased scatter artifacts compared to planning CT (pCT), compromising image quality and limiting further applications. Scatter correction is thus crucial for improving CBCT image quality.

METHODS

In this study, we proposed an unsupervised contrastive learning method for CBCT scatter correction. Initially, we transformed low-quality CBCT into high-quality synthetic pCT (spCT) and generated forward projections of CBCT and spCT. By computing the difference between these projections, we obtained a residual image containing image details and scatter artifacts. Image details primarily comprise high-frequency signals, while scatter artifacts consist mainly of low-frequency signals. We extracted the scatter projection signal by applying a low-pass filter to remove image details. The corrected CBCT (cCBCT) projection signal was obtained by subtracting the scatter artifacts projection signal from the original CBCT projection. Finally, we employed the FDK reconstruction algorithm to generate the cCBCT image.

RESULTS

To evaluate cCBCT image quality, we aligned the CBCT and pCT of six patients. In comparison to CBCT, cCBCT maintains anatomical consistency and significantly enhances CT number, spatial homogeneity, and artifact suppression. The mean absolute error (MAE) of the test data decreased from 88.0623 ± 26.6700 HU to 17.5086 ± 3.1785 HU. The MAE of fat regions of interest (ROIs) declined from 370.2980 ± 64.9730 HU to 8.5149 ± 1.8265 HU, and the error between their maximum and minimum CT numbers decreased from 572.7528 HU to 132.4648 HU. The MAE of muscle ROIs reduced from 354.7689 ± 25.0139 HU to 16.4475 ± 3.6812 HU. We also compared our proposed method with several conventional unsupervised synthetic image generation techniques, demonstrating superior performance.

CONCLUSIONS

Our approach effectively enhances CBCT image quality and shows promising potential for future clinical adoption.

Towards real-time EPID-based 3D in vivo dosimetry for IMRT with Deep Neural Networks: A feasibility study

We investigate the potential of the Deep Dose Estimate (DDE) neural network to predict 3D dose distributions inside patients with Monte Carlo (MC) accuracy, based on transmitted EPID signals and patient CTs. The network was trained using as input patient CTs and first-order dose approximations (FOD). Accurate dose distributions (ADD) simulated with MC were given as training targets. 83 pelvic CTs were used to simulate ADDs and respective EPID signals for subfields of prostate IMRT plans (gantry at 0∘). FODs were produced as backprojections from the EPID signals. 581 ADD-FOD sets were produced and divided into training and test sets. An additional dataset simulated with gantry at 90∘ (lateral set) was used for evaluating the performance of the DDE at different beam directions. The quality of the FODs and DDE-predicted dose distributions (DDEP) with respect to ADDs, from the test and lateral sets, was evaluated with gamma analysis (3%,2 mm). The passing rates between FODs and ADDs were as low as 46%, while for DDEPs the passing rates were above 97% for the test set. Meaningful improvements were also observed for the lateral set. The high passing rates for DDEPs indicate that the DDE is able to convert FODs into ADDs. Moreover, the trained DDE predicts the dose inside a patient CT within 0.6 s/subfield (GPU), in contrast to 14 h needed for MC (CPU-cluster). 3D in vivo dose distributions due to clinical patient irradiation can be obtained within seconds, with MC-like accuracy, potentially paving the way towards real-time EPID-based in vivo dosimetry.

Virtual cone-beam computed tomography simulator with human phantom library and its application to the elemental material decomposition

PURPOSE

The purpose of this study is to develop a virtual CBCT simulator with a head and neck (HN) human phantom library and to demonstrate the feasibility of elemental material decomposition (EMD) for quantitative CBCT imaging using this virtual simulator.

METHODS

The library of 36 HN human phantoms were developed by extending the ICRP 110 adult phantoms based on human age, height, and weight statistics. To create the CBCT database for the library, a virtual CBCT simulator that simulated the direct and scattered X-ray on a flat panel detector using ray-tracing and deep-learning (DL) models was used. Gaussian distributed noise was also included on the flat panel detector, which was evaluated using a real CBCT system. The usefulness of the virtual CBCT system was demonstrated through the application of the developed DL-based EMD model for case involving virtual phantom and real patient.

RESULTS

The virtual simulator could generate various virtual CBCT images based on the human phantom library, and the prediction of the EMD could be successfully performed by preparing the CBCT database from the proposed virtual system, even for a real patient. The CBCT image degradation owing to the scattered X-ray and the statistical noise affected the prediction accuracy, although these effects were minimal. Furthermore, the elemental distribution using the real CBCT image was also predictable.

CONCLUSIONS

This study demonstrated the potential of using computer vision for medical data preparation and analysis, which could have important implications for improving patient outcomes, especially in adaptive radiation therapy.

Image-based scatter correction for cone-beam CT using flip swin transformer U-shape network

BACKGROUND

Cone beam computed tomography (CBCT) plays an increasingly important role in image-guided radiation therapy. However, the image quality of CBCT is severely degraded by excessive scatter contamination, especially in the abdominal region, hindering its further applications in radiation therapy.

PURPOSE

To restore low-quality CBCT images contaminated by scatter signals, a scatter correction algorithm combining the advantages of convolutional neural networks (CNN) and Swin Transformer is proposed.

METHODS

In this paper a scatter correction model for CBCT image, the Flip Swin Transformer U-shape network (FSTUNet) model, is proposed. In this model, the advantages of CNN in texture detail and Swin Transformer in global correlation are used to accurately extract shallow and deep features, respectively. Instead of using the original Swin Transformer tandem structure, we build the Flip Swin Transformer Block to achieve a more powerful inter-window association extraction. The validity and clinical relevance of the method is demonstrated through extensive experiments on a Monte Carlo (MC) simulation dataset and frequency split dataset generated by a validated method, respectively.

RESULT

Experimental results on the MC simulated dataset show that the root mean square error of images corrected by the method is reduced from over 100 HU to about 7 HU. Both the structural similarity index measure (SSIM) and the universal quality index (UQI) are close to 1. Experimental results on the frequency split dataset demonstrate that the method not only corrects shading artifacts but also exhibits a high degree of structural consistency. In addition, comparison experiments show that FSTUNet outperforms UNet, Deep Residual Convolutional Neural Network (DRCNN), DSENet, Pix2pixGAN, and 3DUnet methods in both qualitative and quantitative metrics.

CONCLUSIONS

Accurately capturing the features at different levels is greatly beneficial for reconstructing high-quality scatter-free images. The proposed FSTUNet method is an effective solution to CBCT scatter correction and has the potential to improve the accuracy of CBCT image-guided radiation therapy.

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.

A novel beam stopper-based approach for scatter correction in digital planar radiography

X-ray scatter in planar radiography degrades the contrast resolution of the image, thus reducing its diagnostic utility. Antiscatter grids partially block scattered photons at the cost of increasing the dose delivered by two- to four-fold and posing geometrical restrictions that hinder their use for other acquisition settings, such as portable radiography. The few software-based approaches investigated for planar radiography mainly estimate the scatter map from a low-frequency version of the image. We present a novel method for scatter correction in planar imaging based on direct patient measurements. Samples from the shadowed regions of an additional partially obstructed projection acquired with a beam stopper placed between the X-ray source and the patient are used to estimate the scatter map. Evaluation with simulated and real data showed an increase in contrast resolution for both lung and spine and recovery of ground truth values superior to those of three recently proposed methods. Our method avoids the biases of post-processing methods and yields results similar to those for an antiscatter grid while removing geometrical restrictions at around half the radiation dose. It can be used in unconventional imaging techniques, such as portable radiography, where training datasets needed for deep-learning approaches would be very difficult to obtain.

A correlated sampling-based Monte Carlo simulation for fast CBCT iterative scatter correction

BACKGROUND

In recent years, cone-beam computed tomography (CBCT) has played an important role in medical imaging. However, the applications of CBCT are limited due to the severe scatter contamination. Conventional Monte Carlo (MC) simulation can provide accurate scatter estimation for scatter correction, but the expensive computational cost has always been the bottleneck of MC method in clinical application.

PURPOSE

In this work, an MC simulation method combined with a variance reduction technique called correlated sampling is proposed for fast iterative scatter correction.

METHODS

Correlated sampling exploits correlation between similar simulation systems to reduce the variance of interest quantities. Specifically, conventional MC simulation is first performed on the scatter-contaminated CBCT to generate the initial scatter signal. In the subsequent correction iterations, scatter estimation is then updated by applying correlated MC sampling to the latest corrected CBCT images by reusing the random number sequences of the task-related photons in conventional MC. Afterward, the corrected projections obtained by subtracting the scatter estimation from raw projections are utilized for FDK reconstruction. These steps are repeated until an adequate scatter correction is obtained. The performance of the proposed framework is evaluated by the accuracy of the scatter estimation, the quality of corrected CBCT images and efficiency.

RESULTS

Overall, the difference in mean absolute percentage error between scatter estimation with and without correlated sampling is 0.25% for full-fan case and 0.34% for half-fan case, respectively. In simulation studies, scatter artifacts are substantially eliminated, where the mean absolute error value is reduced from 15 to 2 HU in full-fan case and from 53 to 13 HU in half-fan case. Scatter-to-primary ratio is reduced to 0.02 for full-fan and 0.04 for half-fan, respectively. In phantom study, the contrast-to-noise ratio (CNR) is increased by a factor of 1.63, and the contrast is increased by a factor of 1.77. As for clinical studies, the CNR is improved by 11% and 14% for half-fan and full-fan, respectively. The contrast after correction is increased by 19% for half-fan and 44% for full-fan. Furthermore, root mean square error is also effectively reduced, especially from 78 to 4 HU for full-fan. Experimental results demonstrate that the figure of merit is improved between 23 and 43 folds when using correlated sampling. The proposed method takes less than 25 s for the whole iterative scatter correction process.

CONCLUSIONS

The proposed correlated sampling-based MC simulation method can achieve fast and accurate scatter correction for CBCT, making it suitable for real-time clinical use.

3D-printed large-area focused grid for scatter reduction in cone-beam CT

BACKGROUND

Cone-beam computed tomography (CBCT) systems acquire volumetric data more efficiently than fan-beam or multislice CT, particularly when the anatomy of interest resides within the axial field-of-view of the detector and data can be acquired in one rotation. For such systems, scattered radiation remains a source of image quality degradation leading to increased noise, image artifacts, and CT number inaccuracies.

PURPOSE

Recent advances in metal additive manufacturing allow the production of highly focused antiscatter grids (2D-ASGs) that can be used to reduce scatter intensity, while preserving primary radiation transmission. We present the first implementation of a large-area, 2D-ASG for flat-panel CBCT, including grid-line artifact removal and related improvements in image quality.

METHODS

A 245 × 194 × 10 mm 2D-ASG was manufactured from chrome-cobalt alloy using laser powder-bed fusion (LPBF) (AM-400; Renishaw plc, New Mills Wotton-under-Edge, UK). The 2D-ASG had a square profile with a pitch of 9.09 lines/cm and 10:1 grid-ratio. The nominal 0.1 mm grid septa were focused to a 732 mm x-ray source to optimize primary x-ray transmission and reduce grid-line shadowing at the detector. Powder-bed fusion ensured the structural stability of the ASG with no need for additional interseptal support. The 2D-ASG was coupled to a 0.139-mm element pitch flat-panel detector (DRX 3543, Carestream Health) and proper alignment was confirmed by consistent grid-line shadow thickness across the whole detector array. A 154-mm diameter CBCT image-quality-assurance phantom was imaged using a rotary stage and a ceiling-mounted, x-ray unit (Proteus XR/a, GE Medical Systems, 80kVp, 0.5mAs). Grid-line artifacts were removed using a combination of exposure-dependent gain correction and spatial-frequency, Fourier filtering. Projections were reconstructed using a Parker-weighted, FDK algorithm and voxels were spatially averaged to 357 × 357 × 595 µm to improve the signal-to-noise characteristics of the CBCT reconstruction. Finally, in order to compare image quality with and without scatter, the phantom was scanned again under the same CBCT conditions but with no 2D-ASG. No additional antiscatter (i.e., air-gap, bowtie filtration) strategies were used to evaluate the effects in image quality caused by the 2D-ASG alone.

RESULTS

The large-area, 2D-ASG prototype was successfully designed and manufactured using LPBF. CBCT image-quality improvements using the 2D-ASG included: an overall 14.5% CNR increase across the volume; up to 48.8% CNR increase for low-contrast inserts inside the contrast plate of the QA phantom; and a 65% reduction of cupping artifact in axial profiles of water-filled cross sections of the phantom. Advanced image processing strategies to remove grid line artifacts did not affect the spatial resolution or geometric accuracy of the system.

CONCLUSIONS

LPBF can be used to manufacture highly efficient, 2D-focused ASGs that can be easily coupled to clinical, flat-panel detectors. The implementation of ASGs in CBCT leads to reduced scatter-related artifacts, improved CT number accuracy, and enhanced CNR with no increased equivalent dose to the patient. Further improvements to image quality might be achieved with a combination of scatter-correction algorithms and iterative-reconstruction strategies. Finally, clinical applications where other scatter removal strategies are unfeasible might now achieve superior soft-tissue visualization and quantitative capabilities.

Deep learning for x-ray scatter correction in dedicated breast CT

BACKGROUND

Accurate correction of x-ray scatter in dedicated breast computed tomography (bCT) imaging may result in improved visual interpretation and is crucial to achieve quantitative accuracy during image reconstruction and analysis.

PURPOSE

To develop a deep learning (DL) model to correct for x-ray scatter in bCT projection images.

METHODS

A total of 115 patient scans acquired with a bCT clinical system were segmented into the major breast tissue types (skin, adipose, and fibroglandular tissue). The resulting breast phantoms were divided into training (n = 110) and internal validation cohort (n = 5). Training phantoms were augmented by a factor of four by random translation of the breast in the image field of view. Using a previously validated Monte Carlo (MC) simulation algorithm, 12 primary and scatter bCT projection images with a 30-degree step were generated from each phantom. For each projection, the thickness map and breast location in the field of view were also calculated. A U-Net based DL model was developed to estimate the scatter signal based on the total input simulated image and trained single-projection-wise, with the thickness map and breast location provided as additional inputs. The model was internally validated using MC-simulated projections and tested using an external data set of 10 phantoms derived from images acquired with a different bCT system. For this purpose, the mean relative difference (MRD) and mean absolute error (MAE) were calculated. To test for accuracy in reconstructed images, a full bCT acquisition was mimicked with MC-simulations and then assessed by calculating the MAE and the structural similarity (SSIM). Subsequently, scatter was estimated and subtracted from the bCT scans of three patients to obtain the scatter-corrected image. The scatter-corrected projections were reconstructed and compared with the uncorrected reconstructions by evaluating the correction of the cupping artifact, increase in image contrast, and contrast-to-noise ratio (CNR).

RESULTS

The mean MRD and MAE across all cases (min, max) for the internal validation set were 0.04% (-1.1%, 1.3%) and 2.94% (2.7%, 3.2%), while for the external test set they were -0.64% (-1.6%, 0.2%) and 2.84% (2.3%, 3.5%), respectively. For MC-simulated reconstruction slices, the computed SSIM was 0.99 and the MAE was 0.11% (range: 0%, 0.35%) with a single outlier slice of 2.06%. For the three patient bCT reconstructed images, the correction increased the contrast by a mean of 25% (range: 20%, 30%), and reduced the cupping artifact. The mean CNR increased by 0.32 after scatter correction, which was not found to be significant (95% confidence interval: [-0.01, 0.65], p = 0.059). The time required to correct the scatter in a single bCT projection was 0.2 s on an NVIDIA GeForce GTX 1080 GPU.

CONCLUSION

The developed DL model could accurately estimate scatter in bCT projection images and could enhance contrast and correct for cupping artifact in reconstructed patient images without significantly affecting the CNR. The time required for correction would allow its use in daily clinical practice, and the reported accuracy will potentially allow quantitative reconstructions.

Correction of Bowtie filter induced scatter signals based on air scan data and object scan data

In a cone beam CT system, a bowtie filter brings in additional scatter signals with respect to object induced scatter signals, which can degrade image quality and sometimes result in artifacts. This work aims to improve the image quality of CT scans by analyzing the contribution of bowtie filter induced scatter signals and removing them from projection data. Air calibration is a very useful preprocessing step to eliminate the response variations of detector pixels. Bowtie filter induced scattered x-ray signals of air scans are recorded in air calibration tables and therefore considered as a part of primary signals. However, scattered X-rays behave differently in scanned objects compared to primary x-rays. The difference should be corrected to eliminate the impact of bowtie filter induced scatter signals. A kernel based correction algorithm based on air scan data, named bowtie filter scatter correction algorithm, is applied to estimate and to eliminate the bowtie filter induced scatter signals in object scans. The scatter signals of air scans can be measured with air scans or retrieved from air calibration tables of a CT system, and can be used as input of the correction algorithm to estimate the change of scatter signals caused by the scanned objects in the scan field. Based on the assumption that the scatter signals in the projection data scanned with narrow collimation can be neglected, the difference signals between narrow and broad collimations can be used to estimate bowtie filter induced scatter signals for air scans with the correction of extra-focal radiations (EFRs). The calculated bowtie filter induced scatter signals have been compared with the results of Monte Carlo simulations, and the parameters of correction algorithm have been determined by fitting the measured scatter signal curves of phantom scans with calculated curves. Projection data have been reconstructed using Filtered BackProjection (FBP) method with and without bowtie filter correction to check whether the image quality is improved. Scatter signals can be well approximated with the bowtie filter scatter correction algorithm together with an existing object scatter correction algorithm. After removing the bowtie filter induced scatter signals, the dark bands in reconstructed images in the regions near the edges of scanned objects can be mostly eliminated. The difference signals of air scan data between narrow and broad collimations can be used to estimate the bowtie filter induced scatter for air scans. The proposed bowtie filter scatter correction algorithm using air scan data can be applied to estimate and to remove most of the bowtie filter induced scatter signals in object scans.

Single material beam hardening correction via an analytical energy response model for diagnostic CT

BACKGROUND

Various clinical studies show the potential for a wider quantitative role of diagnostic X-ray computed tomography (CT) beyond size measurements. Currently, the clinical use of attenuation values is however limited due to their lack of robustness. This issue can be observed even on the same scanner across patient size and positioning. There are different causes for the lack of robustness in the attenuation values; one possible source of error is beam hardening of the X-ray source spectrum. The conventional and well-established approach to address this issue is a calibration-based single material beam hardening correction (BHC) using a water cylinder.

PURPOSE

We investigate an alternative approach for single material BHC with the aim of producing a more robust result for the attenuation values. The underlying hypothesis of this investigation is that calibration based BHC automatically corrects for scattered radiation in a manner that is sub-optimal in terms of bias as soon as the scanned object strongly deviates from the water cylinder used for calibration.

METHODS

The approach we propose performs BHC via an analytical energy response model that is embedded into a correction pipeline that efficiently estimates and subtracts scattered radiation in a patient-specific manner prior to BHC. The estimation of scattered radiation is based on minimizing, in average, the squared difference between our corrected data and the vendor-calibrated data. The used energy response model is considering the spectral effects of the detector response and of the pre-filtration of the source spectrum including a beam-shaping bowtie filter. The performance of the correction pipeline is first characterized with computer simulated data. Afterwards, it is tested using real 3-D CT data sets of two different phantoms, with various kV settings and phantom positions, assuming a circular data acquisition. The results are compared in the image domain to those from the scanner.

RESULTS

For experiments with a water cylinder, the proposed correction pipeline leads to similar results as the vendor. For reconstructions of a QRM liver phantom with extension ring, the proposed correction pipeline achieved a more uniform and stable outcome in the attenuation values of homogeneous materials within the phantom. For example, the root mean squared deviation between centered and off-centered phantom positioning was reduced from 6.6 HU to 1.8 HU in one profile.

CONCLUSIONS

We have introduced a patient-specific approach for single material BHC in diagnostic CT via the use of an analytical energy response model. This approach shows promising improvements in terms of robustness of attenuation values for large patient sizes. Our results contribute towards improving CT images so as to make CT attenuation values more reliable for use in clinical practice. This article is protected by copyright. All rights reserved.

Empirical scatter correction (ESC): CBCT scatter artifact reduction without prior information

BACKGROUND

The image quality of cone-beam CT (CBCT) scans severely suffers from scattered radiation if no countermeasures are taken. Scatter artifacts may induce cupping and streak artifacts and lead to a reduced image contrast and wrong CT values of the reconstructed volumes. Established software-based approaches for a correction of scattered radiation typically rely on prior knowledge of the CT system, scan parameters, the scanned object, or all of the aforementioned.

PURPOSE

This study proposes a simple and effective post-processing software-based correction method of scatter artifacts in CBCT scans without specific prior knowledge.

METHODS

We propose the empirical scatter correction (ESC) which generates scatter-like basis images from each projection image by convolution operations. A linear combination of these basis images is subtracted from the original projection image. The logarithm is taken and an FDK reconstruction is performed. The coefficients needed for the linear combination are determined automatically by a downhill simplex algorithm such that the resulting reconstructed images show no scatter artifacts. We demonstrate the potential of ESC by correcting simulated volumes with Monte Carlo scatter artifacts, a head phantom scan performed on our table-top CBCT, and a pelvis scan from a Varian Edge CBCT scanner.

RESULTS

ESC is able to improve the image quality of CBCT scans which is shown on the basis of our simulations and on measured data. For a simulated head CT, the CT value difference to the scatter-free reference image was as low as -6 HU after using ESC whereas the uncorrected data deviated by more than -200 HU from the reference data. Simulations of thorax and abdomen CT scans show that although scatter artifacts are not fully removed, anatomical features which were hard to discover prior to the correction become clearly visible and better segmentable with ESC. Similar results are obtained in the phantom measurement where a comparison to a slit scan of our head phantom shows only small differences. The CT values in soft tissue are improved in this measurement, as well. In soft tissue areas with severe scatter artifacts the CT values agree well with those of the slit scan (difference to slit scan: 35 HU corrected, -289 HU uncorrected). Scatter artifacts in measured patient data can also be reduced using the proposed empirical scatter correction. The results are comparable to those achieved with designated correction algorithms installed on the Varian Edge CBCT system.

CONCLUSIONS

ESC allows to reduce artifacts caused by patient scatter solely based on the projection data. This article is protected by copyright. All rights reserved.

Review of high energy x-ray computed tomography for non-destructive dimensional metrology of large metallic advanced manufactured components

Advanced manufacturing technologies, led by additive manufacturing, have undergone significant growth in recent years. These technologies enable engineers to design parts with reduced weight while maintaining structural and functional integrity. In particular, metal additive manufacturing parts are increasingly used in application areas such as aerospace, where a failure of a mission-critical part can have dire safety consequences. Therefore, the quality of these components is extremely important. A critical aspect of quality control is dimensional evaluation, where measurements provide quantitative results that are traceable to the standard unit of length, the metre. Dimensional measurements allow designers, manufacturers and users to check product conformity against engineering drawings and enable the same quality standard to be used across the supply chain nationally and internationally. However, there is a lack of development of measurement techniques that provide non-destructive dimensional measurements beyond common non-destructive evaluation focused on defect detection. X-ray computed tomography (XCT) technology has great potential to be used as a non-destructive dimensional evaluation technology. However, technology development is behind the demand and growth for advanced manufactured parts. Both the size and the value of advanced manufactured parts have grown significantly in recent years, leading to new requirements of dimensional measurement technologies. This paper is a cross-disciplinary review of state-of-the-art non-destructive dimensional measuring techniques relevant to advanced manufacturing of metallic parts at larger length scales, especially the use of high energy XCT with source energy of greater than 400 kV to address the need in measuring large advanced manufactured parts. Technologies considered as potential high energy x-ray generators include both conventional x-ray tubes, linear accelerators, and alternative technologies such as inverse Compton scattering sources, synchrotron sources and laser-driven plasma sources. Their technology advances and challenges are elaborated on. The paper also outlines the development of XCT for dimensional metrology and future needs.

Adaptive proton therapy

Radiation therapy treatments are typically planned based on a single image set, assuming that the patient's anatomy and its position relative to the delivery system remains constant during the course of treatment. Similarly, the prescription dose assumes constant biological dose-response over the treatment course. However, variations can and do occur on multiple time scales. For treatment sites with significant intra-fractional motion, geometric changes happen over seconds or minutes, while biological considerations change over days or weeks. At an intermediate timescale, geometric changes occur between daily treatment fractions. Adaptive radiation therapy is applied to consider changes in patient anatomy during the course of fractionated treatment delivery. While traditionally adaptation has been done off-line with replanning based on new CT images, online treatment adaptation based on on-board imaging has gained momentum in recent years due to advanced imaging techniques combined with treatment delivery systems. Adaptation is particularly important in proton therapy where small changes in patient anatomy can lead to significant dose perturbations due to the dose conformality and finite range of proton beams. This review summarizes the current state-of-the-art of on-line adaptive proton therapy and identifies areas requiring further research.

CycN-Net: A Convolutional Neural Network Specialized for 4D CBCT Images Refinement

Four-dimensional cone-beam computed tomography (4D CBCT) has been developed to provide a sequence of phase-resolved reconstructions in image-guided radiation therapy. However, 4D CBCT images are degraded by severe streaking artifacts and noise because the phase-resolved image is an extremely sparse-view CT procedure wherein a few under-sampled projections are used for the reconstruction of each phase. Aiming at improving the overall quality of 4D CBCT images, we proposed two CNN models, named N-Net and CycN-Net, respectively, by fully excavating the inherent property of 4D CBCT. To be specific, the proposed N-Net incorporates the prior image reconstructed from entire projection data based on U-Net to boost the image quality for each phase-resolved image. Based on N-Net, a temporal correlation among the phase-resolved images is also considered by the proposed CycN-Net. Extensive experiments on both XCAT simulation data and real patient 4D CBCT datasets were carried out to verify the feasibility of the proposed CNNs. Both networks can effectively suppress streaking artifacts and noise while restoring the distinct features simultaneously, compared with the existing CNN models and two state-of-the-art iterative algorithms. Moreover, the proposed method is robust in handling complicated tasks of various patient datasets and imaging devices, which implies its excellent generalization ability.

Image-based shading correction for narrow-FOV truncated pelvic CBCT with deep convolutional neural networks and transfer learning

Purpose: Cone beam computed tomography (CBCT) is a standard solution for in‐room image guidance for radiation therapy. It is used to evaluate and compensate for anatomopathological changes between the dose delivery plan and the fraction delivery day. CBCT is a fast and versatile solution, but it suffers from drawbacks like low contrast and requires proper calibration to derive density values. Although these limitations are even more prominent with in‐room customized CBCT systems, strategies based on deep learning have shown potential in improving image quality. As such, this article presents a method based on a convolutional neural network and a novel two‐step supervised training based on the transfer learning paradigm for shading correction in CBCT volumes with narrow field of view (FOV) acquired with an ad hoc in‐room system.

Methods: We designed a U‐Net convolutional neural network, trained on axial slices of corresponding CT/CBCT couples. To improve the generalization capability of the network, we exploited two‐stage learning using two distinct data sets. At first, the network weights were trained using synthetic CBCT scans generated from a public data set, and then only the deepest layers of the network were trained again with real‐world clinical data to fine‐tune the weights. Synthetic data were generated according to real data acquisition parameters. The network takes a single grayscale volume as input and outputs the same volume with corrected shading and improved HU values.

Results: Evaluation was carried out with a leave‐one‐out cross‐validation, computed on 18 unique CT/CBCT pairs from six different patients from a real‐world dataset. Comparing original CBCT to CT and improved CBCT to CT, we obtained an average improvement of 6 dB on peak signal‐to‐noise ratio (PSNR), +2% on structural similarity index measure (SSIM). The median interquartile range (IQR) Hounsfield unit (HU) difference between CBCT and CT improved from 161.37 (162.54) HU to 49.41 (66.70) HU. Region of interest (ROI)‐based HU difference was narrowed by 75% in the spongy bone (femoral head), 89% in the bladder, 85% for fat, and 83% for muscle. The improvement in contrast‐to‐noise ratio for these ROIs was about 67%.

Conclusions: We demonstrated that shading correction obtaining CT‐compatible data from narrow‐FOV CBCTs acquired with a customized in‐room system is possible. Moreover, the transfer learning approach proved particularly beneficial for such a shading correction approach.

Slot-scan dual-energy bone densitometry using motorized X-ray systems

PURPOSE

We investigate the feasibility of slot-scan dual-energy (DE) bone densitometry on motorized radiographic equipment. This approach will enable fast quantitative measurements of areal bone mineral density (aBMD) for opportunistic evaluation of osteoporosis.

METHODS

We investigated DE slot-scan protocols to obtain aBMD measurements at the lumbar spine (L-spine) and hip using a motorized x-ray platform capable of synchronized translation of the x-ray source and flat-panel detector (FPD). The slot dimension was 5 × 20 cm² . The DE slot views were processed as follows: (1) convolution kernel-based scatter correction, (2) unfiltered backprojection to tile the slots into long-length radiographs, and (3) projection-domain DE decomposition, consisting of an initial adipose-water decomposition in a bone-free region followed by water-CaHA decomposition with adjustment for adipose content. The accuracy and reproducibility of slot-scan aBMD measurements were investigated using a high-fidelity simulator of a robotic x-ray system (Siemens Multitom Rax) in a total of 48 body phantom realizations: four average bone density settings (cortical bone mass fraction: 10-40%), four body sizes (waist circumference, WC = 70-106 cm), and three lateral shifts of the body within the slot field of view (FOV) (centered and ±1 cm off-center). Experimental validations included: (1) x-ray test-bench feasibility study of adipose-water decomposition and (2) initial demonstration of slot-scan DE bone densitometry on the robotic x-ray system using the European Spine Phantom (ESP) with added attenuation (polymethyl methacrylate [PMMA] slabs) ranging 2 to 6 cm thick.

RESULTS

For the L-spine, the mean aBMD error across all WC settings ranged from 0.08 g/cm² for phantoms with average cortical bone fraction w_cortical  = 10% to ∼0.01 g/cm² for phantoms with w_cortical  = 40%. The L-spine aBMD measurements were fairly robust to changes in body size and positioning, e.g., coefficient of variation (CV) for L1 with w_cortical  = 30% was ∼0.034 for various WC and ∼0.02 for an obese patient (WC = 106 cm) changing lateral shift. For the hip, the mean aBMD error across all phantom configurations was about 0.07 g/cm² for a centered patient. The reproducibility of hip aBMD was slightly worse than in the L-spine (e.g., in the femoral neck, the CV with respect to changing WC was ∼0.13 for phantom realizations with w_cortical  = 30%) due to more challenging scatter estimation in the presence of an air-tissue interface within the slot FOV. The aBMD of the hip was therefore sensitive to lateral positioning of the patient, especially for obese patients: e.g., the CV with respect to patient lateral shift for femoral neck with WC = 106 cm and w_cortical  = 30% was 0.14. Empirical evaluations confirmed substantial reduction in aBMD errors with the proposed adipose estimation procedure and demonstrated robust aBMD measurements on the robotic x-ray system, with aBMD errors of ∼0.1 g/cm² across all three simulated ESP vertebrae and all added PMMA attenuator settings.

CONCLUSIONS

We demonstrated that accurate aBMD measurements can be obtained on a motorized FPD-based x-ray system using DE slot-scans with kernel-based scatter correction, backprojection-based slot view tiling, and DE decomposition with adipose correction.

X-Ray Scatter Estimation Using Deep Splines

X-ray scatter compensation is a very desirable technique in flat-panel X-ray imaging and cone-beam computed tomography. State-of-the-art U-net based scatter removal approaches yielded promising results. However, as there are no physics' constraints applied to the output of the U-Net, it cannot be ruled out that it yields spurious results. Unfortunately, in the context of medical imaging, those may be misleading and could lead to wrong conclusions. To overcome this problem, we propose to embed B-splines as a known operator into neural networks. This inherently constrains their predictions to well-behaved and smooth functions. In a study using synthetic head and thorax data as well as real thorax phantom data, we found that our approach performed on par with U-net when comparing both algorithms based on quantitative performance metrics. However, our approach not only reduces runtime and parameter complexity, but we also found it much more robust to unseen noise levels. While the U-net responded with visible artifacts, the proposed approach preserved the X-ray signal's frequency characteristics.

Comparison of Supervised and Unsupervised Approaches for the Generation of Synthetic CT from Cone-Beam CT

Due to major artifacts and uncalibrated Hounsfield units (HU), cone-beam computed tomography (CBCT) cannot be used readily for diagnostics and therapy planning purposes. This study addresses image-to-image translation by convolutional neural networks (CNNs) to convert CBCT to CT-like scans, comparing supervised to unsupervised training techniques, exploiting a pelvic CT/CBCT publicly available dataset. Interestingly, quantitative results were in favor of supervised against unsupervised approach showing improvements in the HU accuracy (62% vs. 50%), structural similarity index (2.5% vs. 1.1%) and peak signal-to-noise ratio (15% vs. 8%). Qualitative results conversely showcased higher anatomical artifacts in the synthetic CBCT generated by the supervised techniques. This was motivated by the higher sensitivity of the supervised training technique to the pixel-wise correspondence contained in the loss function. The unsupervised technique does not require correspondence and mitigates this drawback as it combines adversarial, cycle consistency, and identity loss functions. Overall, two main impacts qualify the paper: (a) the feasibility of CNN to generate accurate synthetic CT from CBCT images, which is fast and easy to use compared to traditional techniques applied in clinics; (b) the proposal of guidelines to drive the selection of the better training technique, which can be shifted to more general image-to-image translation.

Advances in micro-CT imaging of small animals

PURPOSE

Micron-scale computed tomography (micro-CT) imaging is a ubiquitous, cost-effective, and non-invasive three-dimensional imaging modality. We review recent developments and applications of micro-CT for preclinical research.

METHODS

Based on a comprehensive review of recent micro-CT literature, we summarize features of state-of-the-art hardware and ongoing challenges and promising research directions in the field.

RESULTS

Representative features of commercially available micro-CT scanners and some new applications for both in vivo and ex vivo imaging are described. New advancements include spectral scanning using dual-energy micro-CT based on energy-integrating detectors or a new generation of photon-counting x-ray detectors (PCDs). Beyond two-material discrimination, PCDs enable quantitative differentiation of intrinsic tissues from one or more extrinsic contrast agents. When these extrinsic contrast agents are incorporated into a nanoparticle platform (e.g. liposomes), novel micro-CT imaging applications are possible such as combined therapy and diagnostic imaging in the field of cancer theranostics. Another major area of research in micro-CT is in x-ray phase contrast (XPC) imaging. XPC imaging opens CT to many new imaging applications because phase changes are more sensitive to density variations in soft tissues than standard absorption imaging. We further review the impact of deep learning on micro-CT. We feature several recent works which have successfully applied deep learning to micro-CT data, and we outline several challenges specific to micro-CT.

CONCLUSIONS

All of these advancements establish micro-CT imaging at the forefront of preclinical research, able to provide anatomical, functional, and even molecular information while serving as a testbench for translational research.

Deep learning-based forward and cross-scatter correction in dual-source CT

PURPOSE

Dual-source computed tomography (DSCT) uses two source-detector pairs offset by about 90°. In addition to the well-known forward scatter, a special issue in DSCT is cross-scattered radiation from X-ray tube A detected in the detector of system B and vice versa. This effect can lead to artifacts and reduction of the contrast-to-noise ratio of the images. The purpose of this work is to present and evaluate different deep learning-based methods for scatter correction in DSCT.

METHODS

We present different neural network-based methods for forward and cross-scatter correction in DSCT. These deep scatter estimation (DSE) methods mainly differ in the input and output information that is provided for training and inference and in whether they operate on two-dimensional (2D) or on three-dimensional (3D) data. The networks are trained and validated with scatter distributions obtained by our in-house Monte Carlo simulation. The simulated geometry is adapted to a realistic clinical setup.

RESULTS

All DSE approaches reduce scatter-induced artifacts and lead to superior results than the measurement-based scatter correction. Forward scatter, under the presence of cross-scatter, is best estimated either by our network that uses the current projection and a couple of neighboring views (fDSE 2D few views) or by our 3D network that processes all projections simultaneously (fDSE 3D). Cross-scatter, under the presence of forward scatter, is best estimated using xSSE XDSE 2D, with xSSE referring to a quick single scatter estimate of cross scatter, or by xDSE 3D that uses all projections simultaneously. By using our proposed networks, the total scatter error in dual could be reduced from about 18 HU to approximately 3 HU.

CONCLUSIONS

Deep learning-based scatter correction can reduce scatter artifacts in DSCT. To achieve more accurate cross-scatter estimations, the use of a cross-scatter approximation improves the results. Also, the ability to leverage across different projection angles improves the precision of the algorithm.

A Deep Unsupervised Learning Model for Artifact Correction of Pelvis Cone-Beam CT

Purpose

In recent years, cone-beam computed tomography (CBCT) is increasingly used in adaptive radiation therapy (ART). However, compared with planning computed tomography (PCT), CBCT image has much more noise and imaging artifacts. Therefore, it is necessary to improve the image quality and HU accuracy of CBCT. In this study, we developed an unsupervised deep learning network (CycleGAN) model to calibrate CBCT images for the pelvis to extend potential clinical applications in CBCT-guided ART.

Methods

To train CycleGAN to generate synthetic PCT (sPCT), we used CBCT and PCT images as inputs from 49 patients with unpaired data. Additional deformed PCT (dPCT) images attained as CBCT after deformable registration are utilized as the ground truth before evaluation. The trained uncorrected CBCT images are converted into sPCT images, and the obtained sPCT images have the characteristics of PCT images while keeping the anatomical structure of CBCT images unchanged. To demonstrate the effectiveness of the proposed CycleGAN, we use additional nine independent patients for testing.

Results

We compared the sPCT with dPCT images as the ground truth. The average mean absolute error (MAE) of the whole image on testing data decreased from 49.96 ± 7.21HU to 14.6 ± 2.39HU, the average MAE of fat and muscle ROIs decreased from 60.23 ± 7.3HU to 16.94 ± 7.5HU, and from 53.16 ± 9.1HU to 13.03 ± 2.63HU respectively.

Conclusion

We developed an unsupervised learning method to generate high-quality corrected CBCT images (sPCT). Through further evaluation and clinical implementation, it can replace CBCT in ART.

Artificial intelligence supported single detector multi-energy proton radiography system

Proton radiography imaging was proposed as a promising technique to evaluate internal anatomical changes, to enable pre-treatment patient alignment, and most importantly, to optimize the patient specific CT number to stopping-power ratio conversion. The clinical implementation rate of proton radiography systems is still limited due to their complex bulky design, together with the persistent problem of (in)elastic nuclear interactions and multiple Coulomb scattering (i.e. range mixing). In this work, a compact multi-energy proton radiography system was proposed in combination with an artificial intelligence network architecture (ProtonDSE) to remove the persistent problem of proton scatter in proton radiography. A realistic Monte Carlo model of the Proteus^®One accelerator was built at 200 and 220 MeV to isolate the scattered proton signal in 236 proton radiographies of 80 digital anthropomorphic phantoms. ProtonDSE was trained to predict the proton scatter distribution at two beam energies in a 60%/25%/15% scheme for training, testing, and validation. A calibration procedure was proposed to derive the water equivalent thickness image based on the detector dose response relationship at both beam energies. ProtonDSE network performance was evaluated with quantitative metrics that showed an overall mean absolute percentage error below 1.4% ± 0.4% in our test dataset. For one example patient, detector dose to WET conversions were performed based on the total dose (ITotal), the primary proton dose (IPrimary), and the ProtonDSE corrected detector dose (ICorrected). The determined WET accuracy was compared with respect to the reference WET by idealistic raytracing in a manually delineated region-of-interest inside the brain. The error was determined 4.3% ± 4.1% forWET(ITotal),2.2% ± 1.4% forWET(IPrimary),and 2.5% ± 2.0% forWET(ICorrected).

Evaluation of scatter rejection and correction performance of 2D antiscatter grids in cone beam computed tomography

PURPOSE

We have been investigating two-dimensional (2D) antiscatter grids (2D ASGs) to reduce scatter fluence and improve image quality in cone beam computed tomography (CBCT). In this work, two different aspects of 2D ASGs, their scatter rejection and correction capability, were investigated in CBCT experiments. To correct residual scatter transmitted through the 2D ASG, it was used as a scatter measurement device with a novel method: grid-based scatter sampling.

METHODS

Three focused 2D ASG prototypes with grid ratios of 8, 12, and 16 were developed for linac-mounted offset detector CBCT geometry. In the first phase, 2D ASGs were used as a scatter rejection device, and the effect of grid ratio on CT number accuracy and contrast-to-noise ratio (CNR) evaluated in CBCT images. In the second phase, a grid-based scatter sampling method which exploits the signal modulation characteristics of the 2D ASG's septal shadows to measure and correct residual scatter transmitted through the grid was implemented. To evaluate CT number accuracy, the percent change in CT numbers was measured by changing the phantom from head to pelvis size and configuration.

RESULTS

When 2D ASG was used as a scatter rejection device, CT number accuracy increased and the CT number variation due to change in phantom dimensions was reduced from 23% to 2-6%. A grid ratio of 16 yielded the lowest CT number variation. All three 2D ASGs yielded improvement in CNR, up to a factor of two in pelvis-sized phantoms. When 2D ASG prototypes were used for both scatter rejection and correction, CT number variations were reduced further, to 1.3-2.6%. In comparisons with a clinical CBCT system and a high-performance radiographic ASG, 2D ASG provided higher CT number accuracy under the same imaging conditions.

CONCLUSIONS

When 2D ASG is used solely as a scatter rejection device, substantial improvement in CT number accuracy can be achieved by increasing the grid ratio. Two-dimensional ASGs also provided significant CNR improvement even at lower grid ratios. Two-dimensional ASGs used in conjunction with the grid-based scatter sampling method provided further improvement in CT number accuracy, irrespective of the grid ratio, while preserving 2D ASGs' capacity to improve CNR. The combined effect of scatter rejection and residual scatter correction by 2D ASG may accelerate implementation of new techniques in CBCT that require high quantitative accuracy, such as radiotherapy dose calculation and dual energy CBCT.

An analysis of scatter characteristics in x-ray CT spectral correction

X-ray scatter remains a major physics challenge in volumetric computed tomography (CT), whose physical and statistical behaviors have been commonly leveraged in order to eliminate its impact on CT image quality. In this work, we conduct an in-depth derivation of how the scatter distribution and scatter to primary ratio (SPR) will change during the spectral correction, leading to an interesting finding on the property of scatter. Such a characterization of scatter's behavior provides an analytic approach of compensating for the SPR as well as approximating the change of scatter distribution after spectral correction, even though both of them might be significantly distorted as the linearization mapping function in spectral correction could vary a lot from one detector pixel to another. We conduct an evaluation of SPR compensations on a Catphan phantom and an anthropomorphic chest phantom to validate the characteristics of scatter. In addition, this scatter property is also directly adopted into CT imaging using a spectral modulator with flying focal spot technology (SMFFS) as an example to demonstrate its potential in practical applications. For cone-beam CT scans at both 80 and 120 kVp, CT images with accurate CT numbers can be achieved after spectral correction followed by the appropriate SPR compensation based on our presented scatter property. In the case of the SMFFS based cone-beam CT scan of the Catphan phantom at 120 kVp, after a scatter correction using an analytic algorithm derived from the scatter property, CT image quality was significantly improved, with the averaged root mean square error reduced from 297.9 to 6.5 Hounsfield units (HU).

Comparison of CBCT-based dose calculation methods in head and neck cancer radiotherapy: from Hounsfield unit to density calibration curve to deep learning

PURPOSE

Anatomical variations occur during head and neck (H&N) radiotherapy treatment. kV cone-beam computed tomography (CBCT) images can be used for daily dose monitoring to assess dose variations owing to anatomic changes. Deep learning methods (DLMs) have recently been proposed to generate pseudo-CT (pCT) from CBCT to perform dose calculation. This study aims to evaluate the accuracy of a DLM and to compare this method with three existing methods of dose calculation from CBCT in H&N cancer radiotherapy.

METHODS

Forty-four patients received VMAT for H&N cancer (70-63-56 Gy). For each patient, reference CT (Bigbore, Philips) and CBCT images (XVI, Elekta) were acquired. The DLM was based on a generative adversarial network. The three compared methods were: (a) a method using a density to Hounsfield Unit (HU) relation from phantom CBCT image (HU-D curve method), (b) a water-air-bone density assignment method (DAM), and iii) a method using deformable image registration (DIR). The imaging endpoints were the mean absolute error (MAE) and mean error (ME) of HU from pCT and reference CT (CT_ref ). The dosimetric endpoints were dose discrepancies and 3D gamma analyses (local, 2%/2 mm, 30% dose threshold). Dose discrepancies were defined as the mean absolute differences between DVHs calculated from the CT_ref and pCT of each method.

RESULTS

In the entire body, the MAEs and MEs of the DLM, HU-D curve method, DAM, and DIR method were 82.4 and 17.1 HU, 266.6 and 208.9 HU, 113.2 and 14.2 HU, and 95.5 and -36.6 HU, respectively. The MAE obtained using the DLM differed significantly from those of other methods (Wilcoxon, P ≤ 0.05). The DLM dose discrepancies were 7 ± 8 cGy (maximum = 44 cGy) for the ipsilateral parotid gland D_mean and 5 ± 6 cGy (max = 26 cGy) for the contralateral parotid gland mean dose (D_mean ). For the parotid gland D_mean , no significant dose difference was observed between the DLM and other methods. The mean 3D gamma pass rate ± standard deviation was 98.1 ± 1.2%, 91.0 ± 5.3%, 97.9 ± 1.6%, and 98.8 ± 0.7% for the DLM, HU-D method, DAM, and DIR method, respectively. The gamma pass rates and mean gamma results of the HU-D curve method, DAM, and DIR method differed significantly from those of the DLM.

CONCLUSIONS

For H&N radiotherapy, DIR method and DLM appears as the most appealing CBCT-based dose calculation methods among the four methods in terms of dose accuracy as well as calculation time. Using the DIR method or DLM with CBCT images enables dose monitoring in the parotid glands during the treatment course and may be used to trigger replanning.

What scans we will read: imaging instrumentation trends in clinical oncology

Oncological diseases account for a significant portion of the burden on public healthcare systems with associated costs driven primarily by complex and long-lasting therapies. Through the visualization of patient-specific morphology and functional-molecular pathways, cancerous tissue can be detected and characterized non-invasively, so as to provide referring oncologists with essential information to support therapy management decisions. Following the onset of stand-alone anatomical and functional imaging, we witness a push towards integrating molecular image information through various methods, including anato-metabolic imaging (e.g., PET/CT), advanced MRI, optical or ultrasound imaging.

This perspective paper highlights a number of key technological and methodological advances in imaging instrumentation related to anatomical, functional, molecular medicine and hybrid imaging, that is understood as the hardware-based combination of complementary anatomical and molecular imaging. These include novel detector technologies for ionizing radiation used in CT and nuclear medicine imaging, and novel system developments in MRI and optical as well as opto-acoustic imaging. We will also highlight new data processing methods for improved non-invasive tissue characterization. Following a general introduction to the role of imaging in oncology patient management we introduce imaging methods with well-defined clinical applications and potential for clinical translation. For each modality, we report first on the status quo and, then point to perceived technological and methodological advances in a subsequent status go section. Considering the breadth and dynamics of these developments, this perspective ends with a critical reflection on where the authors, with the majority of them being imaging experts with a background in physics and engineering, believe imaging methods will be in a few years from now.

Overall, methodological and technological medical imaging advances are geared towards increased image contrast, the derivation of reproducible quantitative parameters, an increase in volume sensitivity and a reduction in overall examination time. To ensure full translation to the clinic, this progress in technologies and instrumentation is complemented by advances in relevant acquisition and image-processing protocols and improved data analysis. To this end, we should accept diagnostic images as “data”, and – through the wider adoption of advanced analysis, including machine learning approaches and a “big data” concept – move to the next stage of non-invasive tumour phenotyping. The scans we will be reading in 10 years from now will likely be composed of highly diverse multi-dimensional data from multiple sources, which mandate the use of advanced and interactive visualization and analysis platforms powered by Artificial Intelligence (AI) for real-time data handling by cross-specialty clinical experts with a domain knowledge that will need to go beyond that of plain imaging.

A single neural network for cone-beam computed tomography-based radiotherapy of head-and-neck, lung and breast cancer

•

A deep learning network facilitated dose calculation from CBCT.

•

A single network achieved CBCT-based dose calculation generating synthetic CT for head-and-neck, lung, and breast cancer patients with similar performance to a network specifically trained for each anatomical site.

•

Generation of synthetic-CT can be achieved within 10 s, facilitating online adaptive radiotherapy scenarios.

Background and purpose Adaptive radiotherapy based on cone-beam computed tomography (CBCT) requires high CT number accuracy to ensure accurate dose calculations. Recently, deep learning has been proposed for fast CBCT artefact corrections on single anatomical sites. This study investigated the feasibility of applying a single convolutional network to facilitate dose calculation based on CBCT for head-and-neck, lung and breast cancer patients.

Materials and Methods Ninety-nine patients diagnosed with head-and-neck, lung or breast cancer undergoing radiotherapy with CBCT-based position verification were included in this study. The CBCTs were registered to planning CT according to clinical procedures. Three cycle-consistent generative adversarial networks (cycle-GANs) were trained in an unpaired manner on 15 patients per anatomical site generating synthetic-CTs (sCTs). Another network was trained with all the anatomical sites together. Performances of all four networks were compared and evaluated for image similarity against rescan CT (rCT). Clinical plans were recalculated on rCT and sCT and analysed through voxel-based dose differences and γ-analysis.

Results A sCT was generated in 10 s. Image similarity was comparable between models trained on different anatomical sites and a single model for all sites. Mean dose differences <0.5% were obtained in high-dose regions. Mean gamma (3%, 3 mm) pass-rates >95% were achieved for all sites.

Conclusion Cycle-GAN reduced CBCT artefacts and increased similarity to CT, enabling sCT-based dose calculations. A single network achieved CBCT-based dose calculation generating synthetic CT for head-and-neck, lung, and breast cancer patients with similar performance to a network specifically trained for each anatomical site.

CBCT-based synthetic CT generation using deep-attention cycleGAN for pancreatic adaptive radiotherapy

PURPOSE

Current clinical application of cone-beam CT (CBCT) is limited to patient setup. Imaging artifacts and Hounsfield unit (HU) inaccuracy make the process of CBCT-based adaptive planning presently impractical. In this study, we developed a deep-learning-based approach to improve CBCT image quality and HU accuracy for potential extended clinical use in CBCT-guided pancreatic adaptive radiotherapy.

METHODS

Thirty patients previously treated with pancreas SBRT were included. The CBCT acquired prior to the first fraction of treatment was registered to the planning CT for training and generation of synthetic CT (sCT). A self-attention cycle generative adversarial network (cycleGAN) was used to generate CBCT-based sCT. For the cohort of 30 patients, the CT-based contours and treatment plans were transferred to the first fraction CBCTs and sCTs for dosimetric comparison.

RESULTS

At the site of abdomen, mean absolute error (MAE) between CT and sCT was 56.89 ± 13.84 HU, comparing to 81.06 ± 15.86 HU between CT and the raw CBCT. No significant differences (P > 0.05) were observed in the PTV and OAR dose-volume-histogram (DVH) metrics between the CT- and sCT-based plans, while significant differences (P < 0.05) were found between the CT- and the CBCT-based plans.

CONCLUSIONS

The image similarity and dosimetric agreement between the CT and sCT-based plans validated the dose calculation accuracy carried by sCT. The CBCT-based sCT approach can potentially increase treatment precision and thus minimize gastrointestinal toxicity.

Machine Learning in PET: from Photon Detection to Quantitative Image Reconstruction

Machine learning has found unique applications in nuclear medicine from photon detection to quantitative image reconstruction. While there have been impressive strides in detector development for time-of-flight positron emission tomography, most detectors still make use of simple signal processing methods to extract the time and position information from the detector signals. Now with the availability of fast waveform digitizers, machine learning techniques have been applied to estimate the position and arrival time of high-energy photons. In quantitative image reconstruction, machine learning has been used to estimate various corrections factors, including scattered events and attenuation images, as well as to reduce statistical noise in reconstructed images. Here machine learning either provides a faster alternative to an existing time-consuming computation, such as in the case of scatter estimation, or creates a data-driven approach to map an implicitly defined function, such as in the case of estimating the attenuation map for PET/MR scans. In this article, we will review the abovementioned applications of machine learning in nuclear medicine.

CBCT correction using a cycle-consistent generative adversarial network and unpaired training to enable photon and proton dose calculation

In presence of inter-fractional anatomical changes, clinical benefits are anticipated from image-guided adaptive radiotherapy. Nowadays, cone-beam CT (CBCT) imaging is mostly utilized during pre-treatment imaging for position verification. Due to various artifacts, image quality is typically not sufficient for photon or proton dose calculation, thus demanding accurate CBCT correction, as potentially provided by deep learning techniques. This work aimed at investigating the feasibility of utilizing a cycle-consistent generative adversarial network (cycleGAN) for prostate CBCT correction using unpaired training. Thirty-three patients were included. The network was trained to translate uncorrected, original CBCT images (CBCT_org) into planning CT equivalent images (CBCT_cycleGAN). HU accuracy was determined by comparison to a previously validated CBCT correction technique (CBCT_cor). Dosimetric accuracy was inferred for volumetric-modulated arc photon therapy (VMAT) and opposing single-field uniform dose (OSFUD) proton plans, optimized on CBCT_cor and recalculated on CBCT_cycleGAN. Single-sided SFUD proton plans were utilized to assess proton range accuracy. The mean HU error of CBCT_cycleGAN with respect to CBCT_cor decreased from 24 HU for CBCT_org to  -6 HU. Dose calculation accuracy was high for VMAT, with average pass-rates of 100%/89% for a 2%/1% dose difference criterion. For proton OSFUD plans, the average pass-rate for a 2% dose difference criterion was 80%. Using a (2%, 2 mm) gamma criterion, the pass-rate was 96%. 93% of all analyzed SFUD profiles had a range agreement better than 3 mm. CBCT correction time was reduced from 6-10 min for CBCT_cor to 10 s for CBCT_cycleGAN. Our study demonstrated the feasibility of utilizing a cycleGAN for CBCT correction, achieving high dose calculation accuracy for VMAT. For proton therapy, further improvements may be required. Due to unpaired training, the approach does not rely on anatomically consistent training data or potentially inaccurate deformable image registration. The substantial speed-up for CBCT correction renders the method particularly interesting for adaptive radiotherapy.

Paired cycle-GAN-based image correction for quantitative cone-beam computed tomography

PURPOSE

The incorporation of cone-beam computed tomography (CBCT) has allowed for enhanced image-guided radiation therapy. While CBCT allows for daily 3D imaging, images suffer from severe artifacts, limiting the clinical potential of CBCT. In this work, a deep learning-based method for generating high quality corrected CBCT (CCBCT) images is proposed.

METHODS

The proposed method integrates a residual block concept into a cycle-consistent adversarial network (cycle-GAN) framework, called res-cycle GAN, to learn a mapping between CBCT images and paired planning CT images. Compared with a GAN, a cycle-GAN includes an inverse transformation from CBCT to CT images, which constrains the model by forcing calculation of both a CCBCT and a synthetic CBCT. A fully convolution neural network with residual blocks is used in the generator to enable end-to-end CBCT-to-CT transformations. The proposed algorithm was evaluated using 24 sets of patient data in the brain and 20 sets of patient data in the pelvis. The mean absolute error (MAE), peak signal-to-noise ratio (PSNR), normalized cross-correlation (NCC) indices, and spatial non-uniformity (SNU) were used to quantify the correction accuracy of the proposed algorithm. The proposed method is compared to both a conventional scatter correction and another machine learning-based CBCT correction method.

RESULTS

Overall, the MAE, PSNR, NCC, and SNU were 13.0 HU, 37.5 dB, 0.99, and 0.05 in the brain, 16.1 HU, 30.7 dB, 0.98, and 0.09 in the pelvis for the proposed method, improvements of 45%, 16%, 1%, and 93% in the brain, and 71%, 38%, 2%, and 65% in the pelvis, over the CBCT image. The proposed method showed superior image quality as compared to the scatter correction method, reducing noise and artifact severity. The proposed method produced images with less noise and artifacts than the comparison machine learning-based method.

CONCLUSIONS

The authors have developed a novel deep learning-based method to generate high-quality corrected CBCT images. The proposed method increases onboard CBCT image quality, making it comparable to that of the planning CT. With further evaluation and clinical implementation, this method could lead to quantitative adaptive radiation therapy.

A preliminary study of using a deep convolution neural network to generate synthesized CT images based on CBCT for adaptive radiotherapy of nasopharyngeal carcinoma

This study aims to utilize a deep convolutional neural network (DCNN) for synthesized CT image generation based on cone-beam CT (CBCT) and to apply the images to dose calculations for nasopharyngeal carcinoma (NPC). An encoder-decoder 2D U-Net neural network was produced. A total of 70 CBCT/CT paired images of NPC cancer patients were used for training (50), validation (10) and testing (10) datasets. The testing datasets were treated with the same prescription dose (70 Gy to PTVnx70, 68 Gy to PTVnd68, 62 Gy to the PTV62 and 54 Gy to the PTV54). The mean error (ME) and mean absolute error (MAE) for the true CT images were calculated for image quality evaluation of the synthesized CT. The dose-volume histogram (DVH) dose metric difference and 3D gamma pass rate for the true CT images were calculated for dose analysis, and the results were compared with those for the CBCT images (original CBCT images without any correction) and a patient-specific calibration (PSC) method. Compared with CBCT, the range of the MAE for synthesized CT images improved from (60, 120) to (6, 27) Hounsfield units (HU), and the ME improved from (-74, 51) to (-26, 4) HU. Compared with the true CT method, the average DVH dose metric differences for the CBCT, PSC and synthesized CT methods were 0.8%  ±  1.9%, 0.4%  ±  0.7% and 0.2%  ±  0.6%, respectively. The 1%/1 mm gamma pass rates within the body for the CBCT, PSC and synthesized CT methods were 90.8%  ±  6.2%, 94.1%  ±  4.4% and 95.5%  ±  1.6%, respectively, and the rates within the PTVnx70 were 80.3%  ±  16.6%, 87.9%  ±  19.7%, 98.6%  ±  2.9%, respectively. The DCNN model can generate high-quality synthesized CT images from CBCT images and be used for accurate dose calculations for NPC patients. This finding has great significance for the clinical application of adaptive radiotherapy for NPC.

Comparing Unet training with three different datasets to correct CBCT images for prostate radiotherapy dose calculations

Image intensity correction is crucial to enable cone beam computed tomography (CBCT) based radiotherapy dose calculations. This study evaluated three different deep learning based correction methods using a U-shaped convolutional neural network architecture (Unet) in terms of their photon and proton dose calculation accuracy. CT and CBCT imaging data of 42 prostate cancer patients were included. For target ground truth data generation, a CBCT correction method based on CT to CBCT deformable image registration (DIR) was used. The method yields a deformed CT called (i) virtual CT (vCT) which is used to generate (ii) corrected CBCT projections allowing the reconstruction of (iii) a final corrected CBCT image. The single Unet architecture was trained using these three different datasets: (Unet1) raw and corrected CBCT projections, (Unet2) raw CBCT and vCT image slices and (Unet3) raw and reference corrected CBCT image slices. Volumetric arc therapy (VMAT) and proton pencil beam scanning (PBS) single field uniform dose (SFUD) plans were optimized on the reference corrected image and recalculated on the obtained Unet-corrected CBCT images. The mean error (ME) and mean absolute error (MAE) for Unet1/2/3 were [Formula: see text] Hounsfield units (HU) and [Formula: see text] HU. The 1% dose difference pass rates were better than 98.4% for VMAT for 8 test patients not seen during training, with little difference between Unets. Gamma evaluation results were even better. For protons a gamma evaluation was employed to account for small range shifts, and [Formula: see text] mm pass rates for Unet1/2/3 were better than [Formula: see text] and 91%. A 3 mm range difference threshold was established. Only for Unet3 the 5th and 95th percentiles of the range difference distributions over all fields, test patients and dose profiles were within this threshold. A single Unet architecture was successfully trained using both CBCT projections and CBCT image slices. Since the results of the other Unets were poorer than Unet3, we conclude that training using corrected CBCT image slices as target data is optimal for PBS SFUD proton dose calculations, while for VMAT all Unets provided sufficient accuracy.

Scatter Corrections in X-Ray Computed Tomography: A Physics-Based Analysis

Fundamental limits for the calculation of scattering corrections within X-ray computed tomography (CT) are found within the independent atom approximation from an analysis of the cross sections, CT geometry, and the Nyquist sampling theorem, suggesting large reductions in computational time compared to existing methods. By modifying the scatter by less than 1 %, it is possible to treat some of the elastic scattering in the forward direction as inelastic to achieve a smoother elastic scattering distribution. We present an analysis showing that the number of samples required for the smoother distribution can be greatly reduced. We show that fixed forced detection can be used with many fewer points for inelastic scattering, but that for pure elastic scattering, a standard Monte Carlo calculation is preferred. We use smoothing for both elastic and inelastic scattering because the intrinsic angular resolution is much poorer than can be achieved for projective tomography. Representative numerical examples are given.