Registration-based estimates of local lung tissue expansion compared to xenon CT measures of specific ventilation

Motion and anatomy dual aware lung ventilation imaging by integrating Jacobian map and average CT image using dual path fusion network

BACKGROUND

Deep learning-based computed tomography (CT) ventilation imaging (CTVI) is a promising technique for guiding functional lung avoidance radiotherapy (FLART). However, conventional approaches, which rely on anatomical CT data, may overlook important ventilation features due to the lack of motion data integration.

PURPOSE

This study aims to develop a novel dual-aware CTVI method that integrates both anatomical information from CT images and motional information from Jacobian maps to generate more accurate ventilation images for FLART.

METHODS

A dataset of 66 patients with four-dimensional CT (4DCT) images and reference ventilation images (RefVI) was utilized to develop the dual-path fusion network (DPFN) for synthesizing ventilation images (CTVIDual). The DPFN model was specifically designed to integrate motion data from 4DCT-generated Jacobian maps with anatomical data from average 4DCT images. The DPFN utilized two specialized feature extraction pathways, along with encoders and decoders, designed to handle both 3D average CT images and Jacobian map data. This dual-processing approach enabled the comprehensive extraction of lung ventilation-related features. The performance of DPFN was assessed by comparing CTVIDual to RefVI using various metrics, including Spearman's correlation coefficients (R), Dice similarity coefficients of high-functional region (DSCh), and low-functional region (DSCl). Additionally, CTVIDual was benchmarked against other CTVI methods, including a dual-phase CT-based deep learning method (CTVIDLCT), a radiomics-based method (CTVIFM), a super voxel-based method (CTVISVD), a Unet-based method (CTVIUnet), and two deformable registration-based methods (CTVIJac and CTVIHU).

RESULTS

In the test group, the mean R between CTVIDual and RefVI was 0.70, significantly outperforming CTVIDLCT (0.68), CTVIFM (0.58), CTVISVD (0.62), and CTVIUnet (0.66), with p < 0.05. Furthermore, the DSCh and DSCl values of CTVIDual were 0.64 and 0.80, respectively, outperforming CTVISVD (0.63; 0.73) and CTVIUnet (0.62; 0.77). The performance of CTVIDual was also significantly better than that of CTVIJac and CTVIHU.

CONCLUSIONS

A novel dual-aware CTVI model that integrates anatomical and motion information was developed to synthesize lung ventilation images. It was shown that the accuracy of lung ventilation estimation could be significantly enhanced by incorporating motional information, particularly in patients with tumor-induced blockages. This approach has the potential to improve the accuracy of CTVI, enabling more effective FLART.

Advances in CT-based lung function imaging for thoracic radiotherapy

The objective of this review is to examine the potential benefits and challenges of CT-based lung function imaging in radiotherapy over recent decades. This includes reviewing background information, defining related concepts, classifying and reviewing existing studies, and proposing directions for further investigation. The lung function imaging techniques reviewed herein encompass CT-based methods, specifically utilizing phase-resolved four-dimensional CT (4D-CT) or end-inspiratory and end-expiratory CT scans, to delineate distinct functional regions within the lungs. These methods extract crucial functional parameters, including lung volume and ventilation distribution, pivotal for assessing and characterizing the functional capacity of the lungs. CT-based lung ventilation imaging offers numerous advantages, notably in the realm of thoracic radiotherapy. By utilizing routine CT scans, additional radiation exposure and financial burdens on patients can be avoided. This imaging technique also enables the identification of different functional areas of the lung, which is crucial for minimizing radiation exposure to healthy lung tissue and predicting and detecting lung injury during treatment. In conclusion, CT-based lung function imaging holds significant promise for improving the effectiveness and safety of thoracic radiotherapy. Nevertheless, challenges persist, necessitating further research to address limitations and optimize clinical utilization. Overall, this review highlights the importance of CT-based lung function imaging as a valuable tool in radiotherapy planning and lung injury monitoring.

PixelPrint 4D : A 3D printing method of fabricating patient-specific deformable CT phantoms for respiratory motion applications

All in-vivo medical imaging is impacted by patient motion, especially respiratory motion, which has a significant influence on clinical workflows in diagnostic imaging and radiation therapy. Many technologies such as motion artifact reduction and tumor tracking algorithms have been developed to compensate for respiratory motion during imaging. To assess these technologies, respiratory motion phantoms (RMPs) are required as preclinical testing environments, for instance, in computed tomography (CT). However, current RMPs are highly simplified and do not exhibit realistic tissue structures or deformation patterns. With the rise of more complex motion compensation technologies such as deep learning-based algorithms, there is a need for more realistic RMPs. This work introduces PixelPrint 4D , a 3D printing method designed to fabricate lifelike, patient-specific deformable lung phantoms for CT imaging. The phantom demonstrated accurate replication of patient lung structures, textures, and attenuation profiles. Furthermore, it exhibited accurate nonrigid deformations, volume changes, and attenuation changes under compression. PixelPrint 4D enables the production of highly realistic RMPs, surpassing existing models to offer more robust testing environments for a diverse array of novel CT technologies.

A Phase 2 Randomized Clinical Trial Evaluating 4-Dimensional Computed Tomography Ventilation-Based Functional Lung Avoidance Radiation Therapy for Non-Small Cell Lung Cancer

PURPOSE

To determine whether 4-dimensional computed tomography (4DCT) ventilation-based functional lung avoidance radiation therapy preserves pulmonary function compared with standard radiation therapy for non-small cell lung cancer (NSCLC).

METHODS AND MATERIALS

This single center, randomized, phase 2 trial enrolled patients with NSCLC receiving curative intent radiation therapy with either stereotactic body radiation therapy or conventionally fractionated radiation therapy between 2016 and 2022. Patients were randomized 1:1 to standard of care radiation therapy or functional lung avoidance radiation therapy. The primary endpoint was the change in Jacobian-based ventilation as measured on 4DCT from baseline to 3 months postradiation. Secondary endpoints included changes in volume of high- and low-ventilating lung, pulmonary toxicity, and changes in pulmonary function tests (PFTs).

RESULTS

A total of 122 patients were randomized and 116 were available for analysis. Median follow up was 29.9 months. Functional avoidance plans significantly (P < .05) reduced dose to high-functioning lung without compromising target coverage or organs at risk constraints. When analyzing all patients, there was no difference in the amount of lung showing a reduction in ventilation from baseline to 3 months between the 2 arms (1.91% vs 1.87%; P = .90). Overall grade ≥2 and grade ≥3 pulmonary toxicities for all patients were 24.1% and 8.6%, respectively. There was no significant difference in pulmonary toxicity or changes in PFTs between the 2 study arms. In the conventionally fractionated cohort, there was a lower rate of grade ≥2 pneumonitis (8.2% vs 32.3%; P = .049) and less of a decline in change in forced expiratory volume in 1 second (-3 vs -5; P = .042) and forced vital capacity (1.5 vs -6; P = .005) at 3 months, favoring the functional avoidance arm.

CONCLUSIONS

There was no difference in posttreatment ventilation as measured by 4DCT between the arms. In the cohort of patients treated with conventionally fractionated radiation therapy with functional lung avoidance, there was reduced pulmonary toxicity, and less decline in PFTs suggesting a clinical benefit in patients with locally advanced NSCLC.

Influence of learned landmark correspondences on lung CT registration

BACKGROUND

Disease or injury may cause a change in the biomechanical properties of the lungs, which can alter lung function. Image registration can be used to measure lung ventilation and quantify volume change, which can be a useful diagnostic aid. However, lung registration is a challenging problem because of the variation in deformation along the lungs, sliding motion of the lungs along the ribs, and change in density.

PURPOSE

Landmark correspondences have been used to make deformable image registration robust to large displacements.

METHODS

To tackle the challenging task of intra-patient lung computed tomography (CT) registration, we extend the landmark correspondence prediction model deep convolutional neural network-Match by introducing a soft mask loss term to encourage landmark correspondences in specific regions and avoid the use of a mask during inference. To produce realistic deformations to train the landmark correspondence model, we use data-driven synthetic transformations. We study the influence of these learned landmark correspondences on lung CT registration by integrating them into intensity-based registration as a distance-based penalty.

RESULTS

Our results on the public thoracic CT dataset COPDgene show that using learned landmark correspondences as a soft constraint can reduce median registration error from approximately 5.46 to 4.08 mm compared to standard intensity-based registration, in the absence of lung masks.

CONCLUSIONS

We show that using landmark correspondences results in minor improvements in local alignment, while significantly improving global alignment.

LungViT: Ensembling Cascade of Texture Sensitive Hierarchical Vision Transformers for Cross-Volume Chest CT Image-to-Image Translation

Chest computed tomography (CT) at inspiration is often complemented by an expiratory CT to identify peripheral airways disease. Additionally, co-registered inspiratory-expiratory volumes can be used to derive various markers of lung function. Expiratory CT scans, however, may not be acquired due to dose or scan time considerations or may be inadequate due to motion or insufficient exhale; leading to a missed opportunity to evaluate underlying small airways disease. Here, we propose LungViT- a generative adversarial learning approach using hierarchical vision transformers for translating inspiratory CT intensities to corresponding expiratory CT intensities. LungViT addresses several limitations of the traditional generative models including slicewise discontinuities, limited size of generated volumes, and their inability to model texture transfer at volumetric level. We propose a shifted-window hierarchical vision transformer architecture with squeeze-and-excitation decoder blocks for modeling dependencies between features. We also propose a multiview texture similarity distance metric for texture and style transfer in 3D. To incorporate global information into the training process and refine the output of our model, we use ensemble cascading. LungViT is able to generate large 3D volumes of size 320×320×320 . We train and validate our model using a diverse cohort of 1500 subjects with varying disease severity. To assess model generalizability beyond the development set biases, we evaluate our model on an out-of-distribution external validation set of 200 subjects. Clinical validation on internal and external testing sets shows that synthetic volumes could be reliably adopted for deriving clinical endpoints of chronic obstructive pulmonary disease.

Advances in COPD imaging using CT and MRI: linkage with lung physiology and clinical outcomes

Recent years have witnessed major advances in lung imaging in patients with COPD. These include significant refinements in images obtained by computed tomography (CT) scans together with the introduction of new techniques and software that aim for obtaining the best image whilst using the lowest possible radiation dose. Magnetic resonance imaging (MRI) has also emerged as a useful radiation-free tool in assessing structural and more importantly functional derangements in patients with well-established COPD and smokers without COPD, even before the existence of overt changes in resting physiological lung function tests. Together, CT and MRI now allow objective quantification and assessment of structural changes within the airways, lung parenchyma and pulmonary vessels. Furthermore, CT and MRI can now provide objective assessments of regional lung ventilation and perfusion, and multinuclear MRI provides further insight into gas exchange; this can help in structured decisions regarding treatment plans. These advances in chest imaging techniques have brought new insights into our understanding of disease pathophysiology and characterising different disease phenotypes. The present review discusses, in detail, the advances in lung imaging in patients with COPD and how structural and functional imaging are linked with common resting physiological tests and important clinical outcomes.

Encoding Enhanced Complex CNN for Accurate and Highly Accelerated MRI

Magnetic resonance imaging (MRI) using hyperpolarized noble gases provides a way to visualize the structure and function of human lung, but the long imaging time limits its broad research and clinical applications. Deep learning has demonstrated great potential for accelerating MRI by reconstructing images from undersampled data. However, most existing deep convolutional neural networks (CNN) directly apply square convolution to k-space data without considering the inherent properties of k-space sampling, limiting k-space learning efficiency and image reconstruction quality. In this work, we propose an encoding enhanced (EN2) complex CNN for highly undersampled pulmonary MRI reconstruction. EN2 complex CNN employs convolution along either the frequency or phase-encoding direction, resembling the mechanisms of k-space sampling, to maximize the utilization of the encoding correlation and integrity within a row or column of k-space. We also employ complex convolution to learn rich representations from the complex k-space data. In addition, we develop a feature-strengthened modularized unit to further boost the reconstruction performance. Experiments demonstrate that our approach can accurately reconstruct hyperpolarized 129Xe and 1H lung MRI from 6-fold undersampled k-space data and provide lung function measurements with minimal biases compared with fully sampled images. These results demonstrate the effectiveness of the proposed algorithmic components and indicate that the proposed approach could be used for accelerated pulmonary MRI in research and clinical lung disease patient care.

Systematic Evaluation of the Impact of Lung Segmentation Methods on 4-Dimensional Computed Tomography Ventilation Imaging Using a Large Patient Database

PURPOSE

A novel form of lung functional imaging applied for functional avoidance radiation therapy has been developed that uses 4-dimensional computed tomography (4DCT) data and image processing techniques to calculate lung ventilation (4DCT-ventilation). Lung segmentation is a common step to define a region of interest for 4DCT-ventilation generation. The purpose of this study was to quantitatively evaluate the sensitivity of 4DCT-ventilation imaging using different lung segmentation methods.

METHODS AND MATERIALS

The 4DCT data of 350 patients from 2 institutions were used. Lung contours were generated using 3 methods: (1) reference segmentations that removed airways and pulmonary vasculature manually (Lung-Manual), (2) standard lung contours used for planning (Lung-RadOnc), and (3) artificial intelligence (AI)-based contours that removed the airways and pulmonary vasculature (Lung-AI). The AI model was based on a residual 3-dimensional U-Net and was trained using the Lung-Manual contours of 279 patients. We compared the Lung-RadOnc or Lung-AI with Lung-Manual contours for the entire 4DCT-ventilation functional avoidance process including lung segmentation (surface Dice similarity coefficient [Surface DSC]), 4DCT-ventilation generation (correlation), and subanalysis of 10 patients on a dosimetric endpoint (percentage of high functional volume of lung receiving ≥20 Gy [fV20{%}]).

RESULTS

Surface DSC comparing Lung-Manual/Lung-RadOnc and Lung-Manual/Lung-AI contours was 0.40 ± 0.06 and 0.86 ± 0.04, respectively. The correlation between 4DCT-ventilation images generated with Lung-Manual/Lung-RadOnc and Lung-Manual/Lung-AI were 0.48 ± 0.14 and 0.85 ± 0.14, respectively. The difference in fV20[%] between 4DCT-ventilation generated with Lung-Manual/Lung-RadOnc and Lung-Manual/Lung-AI was 2.5% ± 4.1% and 0.3% ± 0.5%, respectively.

CONCLUSIONS

Our work showed that using standard planning lung contours can result in significantly variable 4DCT-ventilation images. The study demonstrated that AI-based segmentations generate lung contours and 4DCT-ventilation images that are similar to those generated using manual methods. The significance of the study is that it characterizes the lung segmentation sensitivity of the 4DCT-ventilation process and develops methods that can facilitate the integration of this novel imaging in busy clinics.

Robust quantification of CT-ventilation biomarker techniques and repeatability in a porcine model

BACKGROUND

Biomarkers estimating local lung ventilation have been derived from computed tomography (CT) imaging using various image acquisition and post-processing techniques. CT-ventilation biomarkers have potential clinical use in functional avoidance radiation therapy (RT), in which RT treatment plans are optimized to reduce dose delivered to highly ventilated lung. Widespread clinical implementation of CT-ventilation biomarkers necessitates understanding of biomarker repeatability. Performing imaging within a highly controlled experimental design enables quantification of error associated with remaining variables.

PURPOSE

To characterize CT-ventilation biomarker repeatability and dependence on image acquisition and post-processing methodology in anesthetized and mechanically ventilated pigs.

METHODS

Five mechanically ventilated Wisconsin Miniature Swine (WMS) received multiple consecutive four-dimensional CT (4DCT) and maximum inhale and exhale breath-hold CT (BH-CT) scans on five dates to generate CT-ventilation biomarkers. Breathing maneuvers were controlled with an average tidal volume difference <200 cc. As surrogates for ventilation, multiple local expansion ratios (LERs) were calculated from the acquired CT scans using Jacobian-based post-processing techniques.L E R 2 $LER_2$ measured local expansion between an image pair using either inhale and exhale BH-CT images or two 4DCT breathing phase images.L E R N $LER_N$ measured the maximum local expansion across the 4DCT breathing phase images. Breathing maneuver consistency, intra- and interday biomarker repeatability, image acquisition and post-processing technique dependence were quantitatively analyzed.

RESULTS

Biomarkers showed strong agreement with voxel-wise Spearman correlationρ > 0.9 $\rho > 0.9$ for intraday repeatability andρ > 0.8 $\rho > 0.8$ for all other comparisons, including between image acquisition techniques. Intra- and interday repeatability were significantly different (p < 0.01). LER2 and LERN post-processing did not significantly affect intraday repeatability.

CONCLUSIONS

4DCT and BH-CT ventilation biomarkers derived from consecutive scans show strong agreement in controlled experiments with nonhuman subjects.

A hybrid model- and deep learning-based framework for functional lung image synthesis from multi-inflation CT and hyperpolarized gas MRI

BACKGROUND

Hyperpolarized gas MRI is a functional lung imaging modality capable of visualizing regional lung ventilation with exceptional detail within a single breath. However, this modality requires specialized equipment and exogenous contrast, which limits widespread clinical adoption. CT ventilation imaging employs various metrics to model regional ventilation from non-contrast CT scans acquired at multiple inflation levels and has demonstrated moderate spatial correlation with hyperpolarized gas MRI. Recently, deep learning (DL)-based methods, utilizing convolutional neural networks (CNNs), have been leveraged for image synthesis applications. Hybrid approaches integrating computational modeling and data-driven methods have been utilized in cases where datasets are limited with the added benefit of maintaining physiological plausibility.

PURPOSE

To develop and evaluate a multi-channel DL-based method that combines modeling and data-driven approaches to synthesize hyperpolarized gas MRI lung ventilation scans from multi-inflation, non-contrast CT and quantitatively compare these synthetic ventilation scans to conventional CT ventilation modeling.

METHODS

In this study, we propose a hybrid DL configuration that integrates model- and data-driven methods to synthesize hyperpolarized gas MRI lung ventilation scans from a combination of non-contrast, multi-inflation CT and CT ventilation modeling. We used a diverse dataset comprising paired inspiratory and expiratory CT and helium-3 hyperpolarized gas MRI for 47 participants with a range of pulmonary pathologies. We performed six-fold cross-validation on the dataset and evaluated the spatial correlation between the synthetic ventilation and real hyperpolarized gas MRI scans; the proposed hybrid framework was compared to conventional CT ventilation modeling and other non-hybrid DL configurations. Synthetic ventilation scans were evaluated using voxel-wise evaluation metrics such as Spearman's correlation and mean square error (MSE), in addition to clinical biomarkers of lung function such as the ventilated lung percentage (VLP). Furthermore, regional localization of ventilated and defect lung regions was assessed via the Dice similarity coefficient (DSC).

RESULTS

We showed that the proposed hybrid framework is capable of accurately replicating ventilation defects seen in the real hyperpolarized gas MRI scans, achieving a voxel-wise Spearman's correlation of 0.57 ± 0.17 and an MSE of 0.017 ± 0.01. The hybrid framework significantly outperformed CT ventilation modeling alone and all other DL configurations using Spearman's correlation. The proposed framework was capable of generating clinically relevant metrics such as the VLP without manual intervention, resulting in a Bland-Altman bias of 3.04%, significantly outperforming CT ventilation modeling. Relative to CT ventilation modeling, the hybrid framework yielded significantly more accurate delineations of ventilated and defect lung regions, achieving a DSC of 0.95 and 0.48 for ventilated and defect regions, respectively.

CONCLUSION

The ability to generate realistic synthetic ventilation scans from CT has implications for several clinical applications, including functional lung avoidance radiotherapy and treatment response mapping. CT is an integral part of almost every clinical lung imaging workflow and hence is readily available for most patients; therefore, synthetic ventilation from non-contrast CT can provide patients with wider access to ventilation imaging worldwide.

Direct estimation of regional lung volume change from paired and single CT images using residual regression neural network

BACKGROUND

Chest computed tomography (CT) enables characterization of pulmonary diseases by producing high-resolution and high-contrast images of the intricate lung structures. Deformable image registration is used to align chest CT scans at different lung volumes, yielding estimates of local tissue expansion and contraction.

PURPOSE

We investigated the utility of deep generative models for directly predicting local tissue volume change from lung CT images, bypassing computationally expensive iterative image registration and providing a method that can be utilized in scenarios where either one or two CT scans are available.

METHODS

A residual regression convolutional neural network, called Reg3DNet+, is proposed for directly regressing high-resolution images of local tissue volume change (i.e., Jacobian) from CT images. Image registration was performed between lung volumes at total lung capacity (TLC) and functional residual capacity (FRC) using a tissue mass- and structure-preserving registration algorithm. The Jacobian image was calculated from the registration-derived displacement field and used as the ground truth for local tissue volume change. Four separate Reg3DNet+ models were trained to predict Jacobian images using a multifactorial study design to compare the effects of network input (i.e., single image vs. paired images) and output space (i.e., FRC vs. TLC). The models were trained and evaluated on image datasets from the COPDGene study. Models were evaluated against the registration-derived Jacobian images using local, regional, and global evaluation metrics.

RESULTS

Statistical analysis revealed that both factors - network input and output space - were significant determinants for change in evaluation metrics. Paired-input models performed better than single-input models, and model performance was better in the output space of FRC rather than TLC. Mean structural similarity index for paired-input models was 0.959 and 0.956 for FRC and TLC output spaces, respectively, and for single-input models was 0.951 and 0.937. Global evaluation metrics demonstrated correlation between registration-derived Jacobian mean and predicted Jacobian mean: coefficient of determination (r2 ) for paired-input models was 0.974 and 0.938 for FRC and TLC output spaces, respectively, and for single-input models was 0.598 and 0.346. After correcting for effort, registration-derived lobar volume change was strongly correlated with the predicted lobar volume change: for paired-input models r2 was 0.899 for both FRC and TLC output spaces, and for single-input models r2 was 0.803 and 0.862, respectively.

CONCLUSIONS

Convolutional neural networks can be used to directly predict local tissue mechanics, eliminating the need for computationally expensive image registration. Networks that use paired CT images acquired at TLC and FRC allow for more accurate prediction of local tissue expansion compared to networks that use a single image. Networks that only require a single input image still show promising results, particularly after correcting for effort, and allow for local tissue expansion estimation in cases where multiple CT scans are not available. For single-input networks, the FRC image is more predictive of local tissue volume change compared to the TLC image.

Dynamic lung aeration and strain with positive end-expiratory pressure individualized to maximal compliance versus ARDSNet low-stretch strategy: a study in a surfactant depletion model of lung injury

BACKGROUND

Positive end-expiratory pressure (PEEP) individualized to a maximal respiratory system compliance directly implies minimal driving pressures with potential outcome benefits, yet, raises concerns on static and dynamic overinflation, strain and cyclic recruitment. Detailed accurate assessment and understanding of these has been hampered by methodological limitations. We aimed to investigate the effects of a maximal compliance-guided PEEP strategy on dynamic lung aeration, strain and tidal recruitment using current four-dimensional computed tomography (CT) techniques and analytical methods of tissue deformation in a surfactant depletion experimental model of acute respiratory distress syndrome (ARDS).

METHODS

ARDS was induced by saline lung lavage in anesthetized and mechanically ventilated healthy sheep (n = 6). Animals were ventilated in a random sequence with: (1) ARDSNet low-stretch protocol; (2) maximal compliance PEEP strategy. Lung aeration, strain and tidal recruitment were acquired with whole-lung respiratory-gated high-resolution CT and quantified using registration-based techniques.

RESULTS

Relative to the ARDSNet low-stretch protocol, the maximal compliance PEEP strategy resulted in: (1) improved dynamic whole-lung aeration at end-expiration (0.456 ± 0.064 vs. 0.377 ± 0.101, P = 0.019) and end-inspiration (0.514 ± 0.079 vs. 0.446 ± 0.083, P = 0.012) with reduced non-aerated and increased normally-aerated lung mass without associated hyperinflation; (2) decreased aeration heterogeneity at end-expiration (coefficient of variation: 0.498 ± 0.078 vs. 0.711 ± 0.207, P = 0.025) and end-inspiration (0.419 ± 0.135 vs. 0.580 ± 0.108, P = 0.014) with higher aeration in dorsal regions; (3) tidal aeration with larger inspiratory increases in normally-aerated and decreases in poorly-aerated areas, and negligible in hyperinflated lung (Aeration × Strategy: P = 0.026); (4) reduced tidal strains in lung regions with normal-aeration (Aeration × Strategy: P = 0.047) and improved regional distributions with lower tidal strains in middle and ventral lung (Region-of-interest [ROI] × Strategy: P < 0.001); and (5) less tidal recruitment in middle and dorsal lung (ROI × Strategy: P = 0.044) directly related to whole-lung tidal strain (r = 0.751, P = 0.007).

CONCLUSIONS

In well-recruitable ARDS models, a maximal compliance PEEP strategy improved end-expiratory/inspiratory whole-lung aeration and its homogeneity without overinflation. It further reduced dynamic strain in middle-ventral regions and tidal recruitment in middle-dorsal areas. These findings suggest the maximal compliance strategy minimizing whole-lung dynamically quantified mechanisms of ventilator-induced lung injury with less cyclic recruitment and no additional overinflation in large heterogeneously expanded and recruitable lungs.

PhysVENeT: a physiologically-informed deep learning-based framework for the synthesis of 3D hyperpolarized gas MRI ventilation

Functional lung imaging modalities such as hyperpolarized gas MRI ventilation enable visualization and quantification of regional lung ventilation; however, these techniques require specialized equipment and exogenous contrast, limiting clinical adoption. Physiologically-informed techniques to map proton (¹H)-MRI ventilation have been proposed. These approaches have demonstrated moderate correlation with hyperpolarized gas MRI. Recently, deep learning (DL) has been used for image synthesis applications, including functional lung image synthesis. Here, we propose a 3D multi-channel convolutional neural network that employs physiologically-informed ventilation mapping and multi-inflation structural ¹H-MRI to synthesize 3D ventilation surrogates (PhysVENeT). The dataset comprised paired inspiratory and expiratory ¹H-MRI scans and corresponding hyperpolarized gas MRI scans from 170 participants with various pulmonary pathologies. We performed fivefold cross-validation on 150 of these participants and used 20 participants with a previously unseen pathology (post COVID-19) for external validation. Synthetic ventilation surrogates were evaluated using voxel-wise correlation and structural similarity metrics; the proposed PhysVENeT framework significantly outperformed conventional ¹H-MRI ventilation mapping and other DL approaches which did not utilize structural imaging and ventilation mapping. PhysVENeT can accurately reflect ventilation defects and exhibits minimal overfitting on external validation data compared to DL approaches that do not integrate physiologically-informed mapping.

DAART: a deep learning platform for deeply accelerated adaptive radiation therapy for lung cancer

Purpose

The study aimed to implement a novel, deeply accelerated adaptive radiation therapy (DAART) approach for lung cancer radiotherapy (RT). Lung cancer is the most common cause of cancer-related death, and RT is the preferred medically inoperable treatment for early stage non-small cell lung cancer (NSCLC). In the current lengthy workflow, it takes a median of four weeks from diagnosis to RT treatment, which can result in complete restaging and loss of local control with delay. We implemented the DAART approach, featuring a novel deepPERFECT system, to address unwanted delays between diagnosis and treatment initiation.

Materials and methods

We developed a deepPERFECT to adapt the initial diagnostic imaging to the treatment setup to allow initial RT planning and verification. We used data from 15 patients with NSCLC treated with RT to train the model and test its performance. We conducted a virtual clinical trial to evaluate the treatment quality of the proposed DAART for lung cancer radiotherapy.

Results

We found that deepPERFECT predicts planning CT with a mean high-intensity fidelity of 83 and 14 HU for the body and lungs, respectively. The shape of the body and lungs on the synthesized CT was highly conformal, with a dice similarity coefficient (DSC) of 0.91, 0.97, and Hausdorff distance (HD) of 7.9 mm, and 4.9 mm, respectively, compared with the planning CT scan. The tumor showed less conformality, which warrants acquisition of treatment Day1 CT and online adaptive RT. An initial plan was designed on synthesized CT and then adapted to treatment Day1 CT using the adapt to position (ATP) and adapt to shape (ATS) method. Non-inferior plan quality was achieved by the ATP scenario, while all ATS-adapted plans showed good plan quality.

Conclusion

DAART reduces the common online ART (ART) treatment course by at least two weeks, resulting in a 50% shorter time to treatment to lower the chance of restaging and loss of local control.

Validation of CT-based ventilation and perfusion biomarkers with histopathology confirms radiation-induced pulmonary changes in a porcine model

Imaging biomarkers can assess disease progression or prognoses and are valuable tools to help guide interventions. Particularly in lung imaging, biomarkers present an opportunity to extract regional information that is more robust to the patient's condition prior to intervention than current gold standard pulmonary function tests (PFTs). This regional aspect has particular use in functional avoidance radiation therapy (RT) in which treatment planning is optimized to avoid regions of high function with the goal of sparing functional lung and improving patient quality of life post-RT. To execute functional avoidance, detailed dose-response models need to be developed to identify regions which should be protected. Previous studies have begun to do this, but for these models to be clinically translated, they need to be validated. This work validates two metrics that encompass the main components of lung function (ventilation and perfusion) through post-mortem histopathology performed in a novel porcine model. With these methods validated, we can use them to study the nuanced radiation-induced changes in lung function and develop more advanced models.

deepPERFECT: Novel Deep Learning CT Synthesis Method for Expeditious Pancreatic Cancer Radiotherapy

Major sources of delay in the standard of care RT workflow are the need for multiple appointments and separate image acquisition. In this work, we addressed the question of how we can expedite the workflow by synthesizing planning CT from diagnostic CT. This idea is based on the theory that diagnostic CT can be used for RT planning, but in practice, due to the differences in patient setup and acquisition techniques, separate planning CT is required. We developed a generative deep learning model, deepPERFECT, that is trained to capture these differences and generate deformation vector fields to transform diagnostic CT into preliminary planning CT. We performed detailed analysis both from an image quality and a dosimetric point of view, and showed that deepPERFECT enabled the preliminary RT planning to be used for preliminary and early plan dosimetric assessment and evaluation.

Metrics of dose to highly ventilated lung are predictive of radiation-induced pneumonitis in lung cancer patients

PURPOSE

To identify metrics of radiation dose delivered to highly ventilated lung that are predictive of radiation-induced pneumonitis.

METHODS AND MATERIALS

A cohort of 90 patients with locally advanced non-small cell lung cancer treated with standard fractionated radiation therapy (RT) (60-66 Gy in 30-33 fractions) were evaluated. Regional lung ventilation was determined from pre-RT 4-dimensional computed tomography (4DCT) using the Jacobian determinant of a B-spline deformable image registration to estimate lung tissue expansion during respiration. Multiple voxel-wise population- and individual-based thresholds for defining high functioning lung were considered. Mean dose and volumes receiving dose ≥ 5-60 Gy were analyzed for both total lung-ITV (MLD,V5-V60) and highly ventilated functional lung-ITV (fMLD,fV5-fV60). The primary endpoint was symptomatic grade 2+ (G2+) pneumonitis. Receiver operator curve (ROC) analyses were used to identify predictors of pneumonitis.

RESULTS

G2+ pneumonitis occurred in 22.2% of patients, with no differences between stage, smoking status, COPD, or chemo/immunotherapy use between G<2 and G2+ patients (P≥ 0.18). Highly ventilated lung was defined as voxels exceeding the population-wide median of 18% voxel-level expansion. All total and functional metrics were significantly different between patients with and without pneumonitis (P≤ 0.039). Optimal ROC points predicting pneumonitis from functional lung dose were fMLD ≤ 12.3 Gy, fV5 ≤ 54% and fV20 ≤ 19 %. Patients with fMLD ≤ 12.3 Gy had a 14% risk of developing G2+ pneumonitis whereas risk significantly increased to 35% for those with fMLD > 12.3 Gy (P = 0.035).

CONCLUSIONS

Dose to highly ventilated lung is associated with symptomatic pneumonitis and treatment planning strategies should focus on limiting dose to functional regions. These findings provide important metrics to be used in functional lung avoidance RT planning and designing clinical trials.

Four-Dimensional Computed Tomography Ventilation Image-Guided Lung Functional Avoidance Radiation Therapy: A Single-Arm Prospective Pilot Clinical Trial

PURPOSE

The primary objective of this prospective pilot trial was to assess the safety and feasibility of lung functional avoidance radiation therapy (RT) with 4-dimensional (4D) computed tomography (CT) ventilation imaging.

METHODS AND MATERIALS

Patients with primary lung cancer or metastatic disease to the lungs to receive conventionally fractionated RT (CFRT) or stereotactic body RT (SBRT) were eligible. Standard-of-care 4D-CT scans were used to generate ventilation images through image processing/analysis. Each patient required a standard intensity modulated RT plan and ventilation image guided functional avoidance plan. The primary endpoint was the safety of functional avoidance RT, defined as the rate of grade ≥3 adverse events (AEs) that occurred ≤12 months after treatment. Protocol treatment was considered safe if the rates of grade ≥3 pneumonitis and esophagitis were <13% and <21%, respectively for CFRT, and if the rate of any grade ≥3 AEs was <28% for SBRT. Feasibility of functional avoidance RT was assessed by comparison of dose metrics between the 2 plans using the Wilcoxon signed-rank test.

RESULTS

Between May 2015 and November 2019, 34 patients with non-small cell lung cancer were enrolled, and 33 patients were evaluable (n = 24 for CFRT; n = 9 for SBRT). Median follow-up was 14.7 months. For CFRT, the rates of grade ≥3 pneumonitis and esophagitis were 4.2% (95% confidence interval, 0.1%-21.1%) and 12.5% (2.7%-32.4%). For SBRT, no patients developed grade ≥3 AEs. Compared with the standard plans, the functional avoidance plans significantly (P < .01) reduced the lung dose-function metrics without compromising target coverage or adherence to standard organs at risk constraints.

CONCLUSIONS

This study, representing one of the first prospective investigations on lung functional avoidance RT, demonstrated that the 4D-CT ventilation image guided functional avoidance RT that significantly reduced dose to ventilated lung regions could be safely administered, adding to the growing body of evidence for its clinical utility.

Assessment of Regional Lung Ventilation with Positron Emission Tomography Using the Radiofluorinated Gas [18F]SF6: Application to an Animal Model of Impaired Ventilation

PURPOSE

Clinical ventilation studies are primarily performed with computerized tomography (CT) and more often with single-photon emission Computerized tomography (SPECT) using radiolabelled aerosols, both presenting certain limitations. Here, we investigate the use of the radiofluorinated gas [¹⁸F]SF₆ as a positron emission tomography (PET) ventilation marker in an animal model of impaired lung ventilation.

PROCEDURES

Sprague-Dawley rats (n = 15) were randomly assigned to spontaneous ventilation (sham group), endotracheal administration of phosphate-buffered saline (PBS group), or endotracheal administration of lipopolysaccharide (LPS group). PET-[¹⁸F]SF₆ images (10-min acquisition) were acquired at t = 48 h after LPS or PBS administration under mechanical ventilation. CT images were acquired after each PET session. Volumes of interest were manually delineated in the lungs on CT images, and voxel-by-voxel analysis was carried out on PET images to obtain the corresponding histograms. After the imaging sessions, lungs were harvested to conduct histological analysis.

RESULTS

Ventilation studies in sham animals showed uniform distribution of [¹⁸F]SF₆ and fast elimination of the radioactivity after discontinuation of the administration. For PBS- and LPS-treated rats, ventilation defects were observed on PET images in some animals, identified as regions with low presence of the radiolabelled gas. Hypoventilated areas co-localized with regions with higher x-ray attenuation than healthy lungs on the CT images, suggesting the presence of oedema and, in some cases, atelectasis. Histograms obtained from PET images showed quasi-Gaussian distributions for control animals, while PBS- and LPS-treated animals demonstrated the presence of hypoventilated voxels. Deviation of the histograms from Gaussian distribution correlated with histological score was obtained by ex vivo histological analysis.

CONCLUSIONS

[¹⁸F]SF₆ is an appropriate marker of regional lung ventilation and may find application in the early diagnose of acute lung disease.

Quantifying robustness of CT-ventilation biomarkers to image noise

Purpose: To quantify the impact of image noise on CT-based lung ventilation biomarkers calculated using Jacobian determinant techniques. Methods: Five mechanically ventilated swine were imaged on a multi-row CT scanner with acquisition parameters of 120 kVp and 0.6 mm slice thickness in static and 4-dimensional CT (4DCT) modes with respective pitches of 1 and 0.09. A range of tube current time product (mAs) values were used to vary image dose. On two dates, subjects received two 4DCTs: one with 10 mAs/rotation (low-dose, high-noise) and one with CT simulation standard of care 100 mAs/rotation (high-dose, low-noise). Additionally, 10 intermediate noise level breath-hold (BHCT) scans were acquired with inspiratory and expiratory lung volumes. Images were reconstructed with and without iterative reconstruction (IR) using 1 mm slice thickness. The Jacobian determinant of an estimated transformation from a B-spline deformable image registration was used to create CT-ventilation biomarkers estimating lung tissue expansion. 24 CT-ventilation maps were generated per subject per scan date: four 4DCT ventilation maps (two noise levels each with and without IR) and 20 BHCT ventilation maps (10 noise levels each with and without IR). Biomarkers derived from reduced dose scans were registered to the reference full dose scan for comparison. Evaluation metrics were gamma pass rate (Γ) with 2 mm distance-to-agreement and 6% intensity criterion, voxel-wise Spearman correlation (ρ) and Jacobian ratio coefficient of variation (CoV JR ). Results: Comparing biomarkers derived from low (CTDI vol = 6.07 mGy) and high (CTDI vol = 60.7 mGy) dose 4DCT scans, mean Γ, ρ and CoV JR values were 93% ± 3%, 0.88 ± 0.03 and 0.04 ± 0.009, respectively. With IR applied, those values were 93% ± 4%, 0.90 ± 0.04 and 0.03 ± 0.003. Similarly, comparisons between BHCT-based biomarkers with variable dose (CTDI vol = 1.35-7.95 mGy) had mean Γ, ρ and CoV JR of 93% ± 4%, 0.97 ± 0.02 and 0.03 ± 0.006 without IR and 93% ± 4%, 0.97 ± 0.03 and 0.03 ± 0.007 with IR. Applying IR did not significantly change any metrics (p > 0.05). Discussion: This work demonstrated that CT-ventilation, calculated using the Jacobian determinant of an estimated transformation from a B-spline deformable image registration, is invariant to Hounsfield Unit (HU) variation caused by image noise. This advantageous finding may be leveraged clinically with potential applications including dose reduction and/or acquiring repeated low-dose acquisitions for improved ventilation characterization.

Predicting pulmonary ventilation damage after radiation therapy for nonsmall cell lung cancer using a ResNet generative adversarial network

BACKGROUND

Functional lung avoidance radiation therapy (RT) is a technique being investigated to preferentially avoid specific regions of the lung that are predicted to be more susceptible to radiation-induced damage. Reducing the dose delivered to high functioning regions may reduce the occurrence radiation-induced lung injuries (RILIs) and toxicities. However, in order to develop effective lung function-sparing plans, accurate predictions of post-RT ventilation change are needed to determine which regions of the lung should be spared.

PURPOSE

To predict pulmonary ventilation change following RT for nonsmall cell lung cancer using machine learning.

METHODS

A conditional generative adversarial network (cGAN) was developed with data from 82 human subjects enrolled in a randomized clinical trial approved by the institution's IRB to predict post-RT pulmonary ventilation change. The inputs to the network were the pre-RT pulmonary ventilation map and radiation dose distribution. The loss function was a combination of the binary cross-entropy loss and an asymmetrical structural similarity index measure (aSSIM) function designed to increase penalization of under-prediction of ventilation damage. Network performance was evaluated against a previously developed polynomial regression model using a paired sample t-test for comparison. Evaluation was performed using eight-fold cross-validation.

RESULTS

From the eight-fold cross-validation, we found that relative to the polynomial model, the cGAN model significantly improved predicting regions of ventilation damage following radiotherapy based on true positive rate (TPR), 0.14±0.15 to 0.72±0.21, and Dice similarity coefficient (DSC), 0.19±0.16 to 0.46±0.14, but significantly declined in true negative rate, 0.97±0.05 to 0.62±0.21, and accuracy, 0.79±0.08 to 0.65±0.14. Additionally, the average true positive volume increased from 104±119 cc in the POLY model to 565±332 cc in the cGAN model, and the average false negative volume decreased from 654±361 cc in the POLY model to 193±163 cc in the cGAN model.

CONCLUSIONS

The proposed cGAN model demonstrated significant improvement in TPR and DSC. The higher sensitivity of the cGAN model can improve the clinical utility of functional lung avoidance RT by identifying larger volumes of functional lung that can be spared and thus decrease the probability of the patient developing RILIs.

A super-voxel-based method for generating surrogate lung ventilation images from CT

Purpose: This study aimed to develop and evaluate CTVISVD , a super-voxel-based method for surrogate computed tomography ventilation imaging (CTVI). Methods and Materials: The study used four-dimensional CT (4DCT) and single-photon emission computed tomography (SPECT) images and corresponding lung masks from 21 patients with lung cancer obtained from the Ventilation And Medical Pulmonary Image Registration Evaluation dataset. The lung volume of the exhale CT for each patient was segmented into hundreds of super-voxels using the Simple Linear Iterative Clustering (SLIC) method. These super-voxel segments were applied to the CT and SPECT images to calculate the mean density values (D _mean) and mean ventilation values (Vent _mean), respectively. The final CT-derived ventilation images were generated by interpolation from the D _mean values to yield CTVISVD. For the performance evaluation, the voxel- and region-wise differences between CTVISVD and SPECT were compared using Spearman's correlation and the Dice similarity coefficient index. Additionally, images were generated using two deformable image registration (DIR)-based methods, CTVIHU and CTVIJac, and compared with the SPECT images. Results: The correlation between the D _mean and Vent _mean of the super-voxel was 0.59 ± 0.09, representing a moderate-to-high correlation at the super-voxel level. In the voxel-wise evaluation, the CTVISVD method achieved a stronger average correlation (0.62 ± 0.10) with SPECT, which was significantly better than the correlations achieved with the CTVIHU (0.33 ± 0.14, p < 0.05) and CTVIJac (0.23 ± 0.11, p < 0.05) methods. For the region-wise evaluation, the Dice similarity coefficient of the high functional region for CTVISVD (0.63 ± 0.07) was significantly higher than the corresponding values for the CTVIHU (0.43 ± 0.08, p < 0.05) and CTVIJac (0.42 ± 0.05, p < 0.05) methods. Conclusion: The strong correlation between CTVISVD and SPECT demonstrates the potential usefulness of this novel method of ventilation estimation for surrogate ventilation imaging.

4DCT ventilation function image-based functional lung protection for esophageal cancer radiotherapy

BACKGROUND

4DCT (four-dimensional computed tomography) can effectively obtain functional lung ventilation images for patients and integrate them into radiotherapy treatment planning. Studies have not been performed on esophageal cancer, and there is no clear consensus on the optimal functional lung threshold for functional lung.

METHODS

Functional lung images were generated for 11 patients with esophageal cancer. The correlation between the dose-volume parameters of functional lung (FL) as defined by different thresholds and the change of PFT/PDFT (pulmonary [diffusion] function test) metrics before and after radiotherapy were evaluated. FL-sparing planning was generated for each patient to preserve the functional lung and compared to conventional anatomical CT (non-sparing) planning.

RESULTS

There was a significant positive correlation between the FL0.8 (defined Jacobian value ≤ 0.8), FL0.84, and FL0.9 dose-volume parameters and ΔFEV1/FVC (reduction before and after radiotherapy), and the FL0.8‑V₃₀ correlation was the strongest (r = 0.819, P < 0.01). The FL-sparing planning had a target area conformity index and homogeneity index comparable to the non-sparing planning (P > 0.05). For FL, the FL-sparing planning achieved lower FL-MLD (6.30 ± 2.14 Gy vs. 7.83 ± 2.70 Gy), V₁₀ (17.13 ± 7.70% vs. 27.40 ± 9.48%), and V₂₀ (6.96 ± 3.85% vs. 11.63 ± 7.19%) compared to the non-sparing planning (P < 0.05), while heart and spinal cord doses were not significantly different between the two planning groups.

CONCLUSION

The 4DCT-based FL irradiation dose for esophageal cancer was significantly associated with a decrease in FEV1/FVC. The optimal FL defined as a Jacobian value ≤ 0.8 or about 21% of the whole lung volume may be a good choice. FL-sparing planning significantly reduced the FL dose without compromising target area coverage.

Volumetric multiphase ventilation imaging based on four-dimensional computed tomography for functional lung avoidance radiotherapy

PURPOSE

Current computed tomography (CT)-based lung ventilation imaging (CTVI) techniques derive a static ventilation image without temporal information. This research aims to develop a four-dimensional CT (4DCT)-based multiphase dynamic ventilation imaging framework capable of recovering the entire ventilation process throughout the breathing cycle for functional lung avoidance radiotherapy (FLART).

METHODS

A total of 15 free-breathing thoracic 4DCT scans of lung or esophageal cancer patients were collected from the public datasets. The lung region of each phase image was first delineated, and then the mask-free isotropic total variation image registration algorithm was used to derive the deformation vector fields between the end-expiration (EE) phase and other phases. As a surrogate of ventilation, the voxel-wise local expansion ratio of each phase relative to the EE phase was estimated using the parameterized Integrated Jacobian Formulation method in the EE phase coordinate. Lastly, the dynamic ventilation images were generated by warping these phase-specific local expansion distributions with a same geometry into their respective breathing phases. Quantitative analysis, including interphase Spearman correlation coefficients, voxel-wise, and regional-wise expansion/contraction tracking, were performed to indirectly validate the proposed method.

RESULTS

The proposed method maintains the physiological meaning of ventilation on each phase and enables to recover the dynamic lung ventilation process. The mean interphase Spearman correlations ranged between 0.23 ± 0.20 and 0.93 ± 0.04 and decreased near the EE phase. Only 26.2% (2.59E + 6 out of 9.89E + 6) of lung voxels exhibited the same expansion/contraction pattern as the global lung. Qualitative and quantitative evaluations of the interphase ventilation distribution difference show that ventilation spatiotemporal heterogeneities generally exist during respiration.

CONCLUSIONS

In contrast to conventional CTVI metrics, our method enables to extract additional phase-resolved respiration-correlated information and reflects the generally existed ventilation spatiotemporal heterogeneities. Subsequent studies with quantitative phase-by-phase cross-modality evaluations will further explore its potential to deepen our understanding of lung function and respiration mechanics and also to facilitate more accurate implementation of FLART.

Evaluation of tissue shrinkage after CT-guided microwave ablation in patients with liver malignancies using Jacobian determinant

PURPOSE

To assess short-term tissue shrinkage in patients with liver malignancies undergoing computed tomography (CT)-guided microwave ablation (MWA) using Jacobian determinant (JD).

MATERIALS AND METHODS

Twenty-nine patients with 29 hepatic malignancies (primary n = 24; metastases n = 5; median tumor diameter 18 mm) referred to CT-guided MWA (single position; 10 min, 100 W) were included in this retrospective IRB-approved study, after exclusion of five patients. Following segmentation of livers and tumors on pre-interventional images, segmentations were registered on post-interventional images. JD mapping was applied to quantify voxelwise tissue volume changes after MWA. Percentual volume changes were evaluated in the ablated tumor, a 5-cm tumor perimeter and in the whole liver and compared in different clinical conditions (tumor entity: primary vs. secondary; tumor location: subcapsular vs. non-subcapsular; tumor volume: >/<6 ml: cirrhosis: yes vs. no; prior chemotherapy: yes vs. no using Shapiro-Wilk, χ² and Wilcoxon rank sum tests, respectively (with p < 0.05 deemed significant).

RESULTS

Tissue volume change was 0.6% in the ablated tumor, 1.6% in the 5-cm perimeter and 0.3% in the whole liver. Shrinkage in the ablated tumor was pronounced in non-subcapsular located tumors, whereas tissue expansion was noted in subcapsular tumors (median -3.5 vs. 1.1%; p = 0.0195). Shrinkage in the whole liver was higher in tumor volumes >6ml, compared with smaller tumors, in which tissue expansion was noted (median -1.0 vs. 2.5%; p = 0.002). Other clinical conditions had no significant influence on the extent of tissue shrinkage (p > 0.05).

CONCLUSION

3D Jacobian analysis shows that hepatic tissue deformation following MWA is most pronounced in a 5-cm area surrounding the treated tumor. Tumor location and tumor volume may have an impact on the extent of tissue shrinkage which may affect estimation of the safety margin.

[Evaluation of Accuracy of Deformable Image Registration with Newly Developed Treatment Planning Support Software for Thoracic Images]

This study evaluated accuracy of deformable image registration (DIR) with twelve parameter settings for thoracic images. We used peak-inhale and peak-exhale images for ten patients provided by DIR-lab. We used a prototype version of iCView software (ITEM Corporation) with DIR to perform intensity, structure, and hybrid-based DIR with the twelve parameter settings. DIR accuracy was evaluated by a target registration error (TRE) using 300 bronchial bifurcations and the Dice similarity coefficient (DSC) of the lungs. For twelve parameter settings, TRE ranged from 2.83 mm to 5.27 mm, whereas DSC ranged from 0.96 to 0.98. These results demonstrated that DIR accuracy differed among parameter settings and show that appropriate parameter settings are required for clinical practice.

Different functional lung-sparing strategies and radiotherapy techniques for patients with esophageal cancer

Background

Integration of 4D-CT ventilation function images into esophageal cancer radiation treatment planning aimed to assess dosimetric differences between different functional lung (FL) protection strategies and radiotherapy techniques.

Methods

A total of 15 patients with esophageal cancer who had 4D-CT scans were included. Lung ventilation function images based on Jacobian values were obtained by deformation image registration and ventilation imaging algorithm. Several different plans were designed for each patient: clinical treatment planning (non-sparing planning), the same beam distribution to FL-sparing planning, three fixed-beams FL-sparing intensity-modulated radiation therapy (IMRT) planning (5F-IMRT, 7F-IMRT, 9F-IMRT), and two FL-sparing volumetric modulated arc therapy (VMAT) planning [1F-VMAT (1-Arc), 2F-VMAT (2-Arc)]. The dosimetric parameters of the planning target volume (PTV) and organs at risk (OARs) were compared and focused on dosimetric differences in FL.

Results

The FL-sparing planning compared with the non-sparing planning significantly decreased the FL-Dmean, V5-30 and Lungs-Dmean, V10-30 (Vx: volume of receiving ≥X Gy), although it slightly compromised PTV conformability and increased Heart-V40 (P&lt; 0.05). The 5F-IMRT had the lowest PTV-conformability index (CI) but had a lower Lungs and Heart irradiation dose compared with those of the 7F-IMRT and 9F-IMRT (P&lt; 0.05). The 2F-VMAT had higher PTV-homogeneity index (HI) and reduced irradiation dose to FL, Lungs, and Heart compared to those of the 1F-VMAT planning (P&lt; 0.05). The 2F-VMAT had higher PTV conformability and homogeneity and decreased FL-Dmean, V5-20 and Lungs-Dmean, V5-10 but correspondingly increased spinal cord-Dmean compared with those of the 5F-IMRT planning (P&lt; 0.05).

Conclusion

In this study, 4D-CT ventilation function image-based FL-sparing planning for esophageal cancer can effectively reduce the dose of the FL. The 2F-VMAT planning is better than the 5F-IMRT planning in reducing the dose of FL.

Innovations in thoracic imaging: CT, radiomics, AI and x-ray velocimetry

In recent years, pulmonary imaging has seen enormous progress, with the introduction, validation and implementation of new hardware and software. There is a general trend from mere visual evaluation of radiological images to quantification of abnormalities and biomarkers, and assessment of ‘non visual’ markers that contribute to establishing diagnosis or prognosis. Important catalysts to these developments in thoracic imaging include new indications (like computed tomography [CT] lung cancer screening) and the COVID‐19 pandemic. This review focuses on developments in CT, radiomics, artificial intelligence (AI) and x‐ray velocimetry for imaging of the lungs. Recent developments in CT include the potential for ultra‐low‐dose CT imaging for lung nodules, and the advent of a new generation of CT systems based on photon‐counting detector technology. Radiomics has demonstrated potential towards predictive and prognostic tasks particularly in lung cancer, previously not achievable by visual inspection by radiologists, exploiting high dimensional patterns (mostly texture related) on medical imaging data. Deep learning technology has revolutionized the field of AI and as a result, performance of AI algorithms is approaching human performance for an increasing number of specific tasks. X‐ray velocimetry integrates x‐ray (fluoroscopic) imaging with unique image processing to produce quantitative four dimensional measurement of lung tissue motion, and accurate calculations of lung ventilation.

See relatedEditorial

A Transfer Learning Framework for Deep Learning-Based CT-to-Perfusion Mapping on Lung Cancer Patients

Purpose

Deep learning model has shown the feasibility of providing spatial lung perfusion information based on CT images. However, the performance of this method on lung cancer patients is yet to be investigated. This study aims to develop a transfer learning framework to evaluate the deep learning based CT-to-perfusion mapping method specifically on lung cancer patients.

Methods

SPECT/CT perfusion scans of 33 lung cancer patients and 137 non-cancer patients were retrospectively collected from two hospitals. To adapt the deep learning model on lung cancer patients, a transfer learning framework was developed to utilize the features learned from the non-cancer patients. These images were processed to extract features from three-dimensional CT images and synthesize the corresponding CT-based perfusion images. A pre-trained model was first developed using a dataset of patients with lung diseases other than lung cancer, and subsequently fine-tuned specifically on lung cancer patients under three-fold cross-validation. A multi-level evaluation was performed between the CT-based perfusion images and ground-truth SPECT perfusion images in aspects of voxel-wise correlation using Spearman’s correlation coefficient (R), function-wise similarity using Dice Similarity Coefficient (DSC), and lobe-wise agreement using mean perfusion value for each lobe of the lungs.

Results

The fine-tuned model yielded a high voxel-wise correlation (0.8142 ± 0.0669) and outperformed the pre-trained model by approximately 8%. Evaluation of function-wise similarity indicated an average DSC value of 0.8112 ± 0.0484 (range: 0.6460-0.8984) for high-functional lungs and 0.8137 ± 0.0414 (range: 0.6743-0.8902) for low-functional lungs. Among the 33 lung cancer patients, high DSC values of greater than 0.7 were achieved for high functional volumes in 32 patients and low functional volumes in all patients. The correlations of the mean perfusion value on the left upper lobe, left lower lobe, right upper lobe, right middle lobe, and right lower lobe were 0.7314, 0.7134, 0.5108, 0.4765, and 0.7618, respectively.

Conclusion

For lung cancer patients, the CT-based perfusion images synthesized by the transfer learning framework indicated a strong voxel-wise correlation and function-wise similarity with the SPECT perfusion images. This suggests the great potential of the deep learning method in providing regional-based functional information for functional lung avoidance radiation therapy.

Computed Tomography Registration-Derived Regional Ventilation Indices Compared to Global Lung Function Parameters in Patients With COPD

CT registration-derived indices provide data on regional lung functional changes in COPD. However, because unlike spirometry which involves dynamic maximal breathing maneuvers, CT-based functional parameters are assessed between two static breath-holds, it is not clear how regional and global lung function parameters relate to each other. We assessed the relationship between CT-density change (dHU), specific volume change (dsV), and regional lung tissue deformation (J) with global spirometric and plethysmographic parameters, gas exchange, exercise capacity, dyspnoea, and disease stage in a prospective cohort study in 102 COPD patients. There were positive correlations of dHU, dsV, and J with spirometric variables, DLCO and gas exchange, 6-min walking distance, and negative correlations with plethysmographic lung volumes and indices of trapping and lung distension as well as GOLD stage. Stepwise regression identified FEV1/FVC (standardized β = 0.429, p &lt; 0.0001), RV/TLC (β = −0.37, p &lt; 0.0001), and BMI (β = 0.27, p=&lt;0.001) as the strongest predictors of CT intensity-based metrics dHU, with similar findings for dsV, while FEV1/FVC (β = 0.32, p=&lt;0.001) and RV/TLC (β = −0.48, p=&lt;0.0001) were identified as those for J. These data suggest that regional lung function is related to two major pathophysiological processes involved in global lung function deterioration in COPD: chronic airflow obstruction and gas trapping, with an additional contribution of nutritional status, which in turn determines respiratory muscle strength. Our data confirm previous findings in the literature, suggesting the potential of CT image-based regional lung function metrics as the biomarkers of disease severity and provide mechanistic insight into the interpretation of regional lung function indices in patients with COPD.

Deriving Pulmonary Ventilation Images From Clinical 4D-CBCT Using a Deep Learning-Based Model

Purpose

The current algorithms for measuring ventilation images from 4D cone-beam computed tomography (CBCT) are affected by the accuracy of deformable image registration (DIR). This study proposes a new deep learning (DL) method that does not rely on DIR to derive ventilation images from 4D-CBCT (CBCT-VI), which was validated with the gold-standard single-photon emission-computed tomography ventilation image (SPECT-VI).

Materials and Methods

This study consists of 4D-CBCT and 99mTc-Technegas SPECT/CT scans of 28 esophagus or lung cancer patients. The scans were rigidly registered for each patient. Using these data, CBCT-VI was derived using a deep learning-based model. Two types of model input data are studied, namely, (a) 10 phases of 4D-CBCT and (b) two phases of peak-exhalation and peak-inhalation of 4D-CBCT. A sevenfold cross-validation was applied to train and evaluate the model. The DIR-dependent methods (density-change-based and Jacobian-based methods) were used to measure the CBCT-VIs for comparison. The correlation was calculated between each CBCT-VI and SPECT-VI using voxel-wise Spearman's correlation. The ventilation images were divided into high, medium, and low functional lung regions. The similarity of different functional lung regions between SPECT-VI and each CBCT-VI was evaluated using the dice similarity coefficient (DSC). One-factor ANONA model was used for statistical analysis of the averaged DSC for the different methods of generating ventilation images.

Results

The correlation values were 0.02 ± 0.10, 0.02 ± 0.09, and 0.65 ± 0.13/0.65 ± 0.15, and the averaged DSC values were 0.34 ± 0.04, 0.34 ± 0.03, and 0.59 ± 0.08/0.58 ± 0.09 for the density change, Jacobian, and deep learning methods, respectively. The strongest correlation and the highest similarity with SPECT-VI were observed for the deep learning method compared to the density change and Jacobian methods.

Conclusion

The results showed that the deep learning method improved the accuracy of correlation and similarity significantly, and the derived CBCT-VIs have the potential to monitor the lung function dynamic changes during radiotherapy.

Artificial intelligence in functional imaging of the lung

Artificial intelligence (AI) is transforming the way we perform advanced imaging. From high-resolution image reconstruction to predicting functional response from clinically acquired data, AI is promising to revolutionize clinical evaluation of lung performance, pushing the boundary in pulmonary functional imaging for patients suffering from respiratory conditions. In this review, we overview the current developments and expound on some of the encouraging new frontiers. We focus on the recent advances in machine learning and deep learning that enable reconstructing images, quantitating, and predicting functional responses of the lung. Finally, we shed light on the potential opportunities and challenges ahead in adopting AI for functional lung imaging in clinical settings.

Origins of and lessons from quantitative functional X-ray computed tomography of the lung

Functional CT of the lung has emerged from quantitative CT (qCT). Structural details extracted at multiple lung volumes offer indices of function. Additionally, single volumetric images, if acquired at standardized lung volumes and body posture, can be used to model function by employing such engineering techniques as computational fluid dynamics. With the emergence of multispectral CT imaging including dual energy from energy integrating CT scanners and multienergy binning using the newly released photon counting CT technology, function is tagged via use of contrast agents. Lung disease phenotypes have previously been lumped together by the limitations of spirometry and plethysmography. QCT and its functional embodiment have been imbedded into studies seeking to characterize chronic obstructive pulmonary disease, severe asthma, interstitial lung disease and more. Reductions in radiation dose by an order of magnitude or more have been achieved. At the same time, we have seen significant increases in spatial and density resolution along with methodologic validations of extracted metrics. Together, these have allowed attention to turn towards more mild forms of disease and younger populations. In early applications, clinical CT offered anatomic details of the lung. Functional CT offers regional measures of lung mechanics, the assessment of functional small airways disease, as well as regional ventilation-perfusion matching (V/Q) and more. This paper will focus on the use of quantitative/functional CT for the non-invasive exploration of dynamic three-dimensional functioning of the breathing lung and beating heart within the unique negative pressure intrathoracic environment of the closed chest.

Joint Deformable Image Registration and ADC Map Regularization: Application to DWI-based Lymphoma Classification

The Apparent Diffusion Coefficient (ADC) is considered an important imaging biomarker contributing to the assessment of tissue microstructure and pathophysiology. It is calculated from Diffusion-Weighted Magnetic Resonance Imaging (DWI) by means of a diffusion model, usually without considering any motion during image acquisition. We propose a method to improve the computation of the ADC by coping jointly with both motion artifacts in whole-body DWI (through group-wise registration) and possible instrumental noise in the diffusion model. The proposed deformable registration method yielded on average the lowest ADC reconstruction error on data with simulated motion and diffusion. Moreover, our approach was applied on whole-body diffusion weighted images obtained with five different b-values from a cohort of 38 patients with histologically confirmed lymphomas of three different types (Hodgkin, diffuse large B-cell lymphoma and follicular lymphoma). Evaluation on the real data showed that ADC-based features, extracted using our joint optimization approach classified lymphomas with an accuracy of approximately 78.6\% (yielding a 11\% increase in respect to the standard features extracted from unregistered diffusion-weighted images). Furthermore, the correlation between diffusion characteristics and histopathological findings was higher than any other previous approach of ADC computation.

Prediction of radiation pneumonitis with machine learning using 4D-CT based dose-function features

In this article, we highlight the fundamental importance of the simultaneous use of dose-volume histogram (DVH) and dose-function histogram (DFH) features based on functional images calculated from 4-dimensional computed tomography (4D-CT) and deformable image registration (DIR) in developing a multivariate radiation pneumonitis (RP) prediction model. The patient characteristics, DVH features and DFH features were calculated from functional images by Hounsfield unit (HU) and Jacobian metrics, for an RP grade ≥ 2 multivariate prediction models were computed from 85 non-small cell lung cancer patients. The prediction model is developed using machine learning via a kernel-based support vector machine (SVM) machine. In the patient cohort, 21 of the 85 patients (24.7%) presented with RP grade ≥ 2. The median area under curve (AUC) was 0.58 for the generated 50 prediction models with patient clinical features and DVH features. When HU metric and Jacobian metric DFH features were added, the AUC improved to 0.73 and 0.68, respectively. We conclude that predictive RP models that incorporate DFH features were successfully developed via kernel-based SVM. These results demonstrate that effectiveness of the simultaneous use of DVH features and DFH features calculated from 4D-CT and DIR on functional image-guided radiotherapy.

Predicting radiation pneumonitis with fuzzy clustering neural network using 4DCT ventilation image based dosimetric parameters

Background

To develop a fuzzy clustering neural network to predict radiation-induced pneumonitis (RP) using four-dimensional computed tomography (4DCT) ventilation image (VI) based dosimetric parameters for thoracic cancer patients.

Methods

The VI were retrospectively calculated from pre-treatment 4DCT data using a deformable image registration (DIR) and an improved VI algorithm. Similar to dose-volume histogram (DVH) of intensity modulated radiotherapy (IMRT), dose-function histogram (DFH) was derived from dose distribution and VI. Then, the dose-function metrics were calculated from DFH. For comparison, the dose-volume metrics were calculated from DVH. Correspondingly, two sets of feature vectors were formed from the dose-volume metrics and the dose-function metrics, respectively. For the feature vectors of each set, they were first pre-processed by principal component analysis (PCA) to reduce feature dimensions. Then, they were grouped to few clusters determined by fuzzy c-means (FCM) algorithm. Next, the neural network was trained to correlate the dosimetric parameters with RP based on the feature vectors of each cluster. Finally, the occurrence of RP was predicted by the neural network on the test data.

Results

Through PCA analysis, the top 5 principal components were selected. Their contribution is more than 98%, which is adequate to represent the original feature space of input data. Based on the clustering validity indexes, the optimal number of clusters is 4 and used for subsequent fuzzy clustering of the input data. After network training, the areas under the curve (AUC) of the prediction model is 0.77 using VI-based dosimetric parameters and 0.67 using structure-based dosimetric parameters.

Conclusions

Compared to the structure-based dosimetric features, the VI-based dosimetric features are more relevant to lung function and presented higher prediction accuracy of RP. The fuzzy clustering neural network improved the prediction accuracy of RP compared to the conventional neural network. The combination of VI-based dose-function metrics and fuzzy clustering neural network provides an effective predictive model for assessing lung toxicity risk after radiotherapy.

Lung-CRNet: A convolutional recurrent neural network for lung 4DCT image registration

PURPOSE

Deformable image registration (DIR) of lung four-dimensional computed tomography (4DCT) plays a vital role in a wide range of clinical applications. Most of the existing deep learning-based lung 4DCT DIR methods focus on pairwise registration which aims to register two images with large deformation. However, the temporal continuities of deformation fields between phases are ignored. This paper proposes a fast and accurate deep learning-based lung 4DCT DIR approach that leverages the temporal component of 4DCT images.

METHODS

We present Lung-CRNet, an end-to-end convolutional recurrent registration neural network for lung 4DCT images and reformulate 4DCT DIR as a spatiotemporal sequence predicting problem in which the input is a sequence of three-dimensional computed tomography images from the inspiratory phase to the expiratory phase in a respiratory cycle. The first phase in the sequence is selected as the only reference image and the rest as moving images. Multiple convolutional gated recurrent units (ConvGRUs) are stacked to capture the temporal clues between images. The proposed network is trained in an unsupervised way using a spatial transformer layer. During inference, Lung-CRNet is able to yield the respective displacement field for each reference-moving image pair in the input sequence.

RESULTS

We have trained the proposed network using a publicly available lung 4DCT dataset and evaluated performance on the widely used the DIR-Lab dataset. The mean and standard deviation of target registration error are 1.56 ± 1.05 mm on the DIR-Lab dataset. The computation time for each forward prediction is less than 1 s on average.

CONCLUSIONS

The proposed Lung-CRNet is comparable to the existing state-of-the-art deep learning-based 4DCT DIR methods in both accuracy and speed. Additionally, the architecture of Lung-CRNet can be generalized to suit other groupwise registration tasks which align multiple images simultaneously.

Dynamic lung compliance imaging from 4DCT-derived volume change estimation

Background. Lung compliance (LC) is the ability of the lung to expand with changes in pressure and is one of the earliest physiological measurements to be altered in patients with parenchymal lung disease. Therefore, compliance monitoring could potentially identify patients at risk for disease progression. However, in clinical practice, compliance measurements are prohibitively invasive for use as a routine monitoring tool.Purpose. We propose a novel method for computing dynamic lung compliance imaging (LCI) from non-contrast computed tomography (CT) scans. LCI applies image processing methods to free-breathing 4DCT images, acquired under two different continuous positive airway pressures (CPAP) applied using a full-face mask, in order to compute the lung volume change induced by the pressure change. LCI provides a quantitative volumetric map of lung stiffness.Methods. We compared mean LCI values computed for 10 patients with idiopathic pulmonary fibrosis (IPF) and 7 non-IPF patients who were screened for lung nodules. 4DCTs were acquired for each patient at 5 cm and 10 cm H₂0 CPAP, as the patients were free breathing at functional residual capacity. LCI was computed from the two 4DCTs. Mean LCI intensities, which represent relative voxel volume change induced by the change in CPAP pressure, were computed.Results.The mean LCI values for patients with IPF ranged between [0.0309, 0.1165], whereas the values ranged between [0.0704, 0.2185] for the lung nodule cohort. Two-sided Wilcoxon rank sum test indicated that the difference in medians is statistically significant (pvalue = 0.009) and that LCI -measured compliance is overall lower in the IPF patient cohort.Conclusion. There is considerable difference in LC scores between patients with IPF compared to controls. Future longitudinal studies should look for LC alterations in areas of lung prior to radiographic detection of fibrosis to further characterize LCI's potential utility as an image marker for disease progression.

Identification of modes of tumor regression in non-small cell lung cancer patients during radiotherapy

PURPOSE

Observed gross tumor volume (GTV) shrinkage during radiotherapy (RT) raises the question of whether to adapt treatment to changes observed on the acquired images. In the literature, two modes of tumor regression have been described: elastic and non-elastic. These modes of tumor regression will affect the safety of treatment adaptation. This study applies a novel approach, using routine cone-beam computed tomography (CBCT) and deformable image registration to automatically distinguish between elastic and non-elastic tumor regression.

METHODS

In this retrospective study, 150 locally advanced non-small cell lung cancer patients treated with 55 Gray of radiotherapy were included. First, the two modes of tumor regression were simulated. For each mode of tumor regression, one timepoint was simulated. Based on the results of simulated data, the approach used for analysis in real patients was developed. CBCTs were non-rigidly registered to the baseline CBCT using a cubic B-spline algorithm, NiftyReg. Next, the Jacobian determinants were computed from the deformation vector fields. To capture local volume changes, 10 Jacobian values were sampled perpendicular to the surface of the GTV, across the lung-tumor boundary. From the simulated data, we can distinguish elastic from non-elastic tumor regression by comparing the Jacobian values samples between 5 and 12.5 mm inside and 5 and 12.5 mm outside the planning GTV. Finally, morphometric results were compared between tumors of different histologies.

RESULTS

Most patients (92.3%) in our cohort showed stable disease in the first week of treatment and non-elastic shrinkage in the later weeks of treatment. At week 2, 125 patients (88%) showed stable disease, three patients (2.1%) disease progression, and 11 patients (8%) regression. By treatment completion, 91 patients (64%) had stable disease, one patient (0.7%) progression and 46 patients (32%) regression. A slight difference in the mode of tumor change was observed between tumors of different histologies.

CONCLUSION

Our novel approach shows that it may be possible to automatically quantify and identify global changes in lung cancer patients during RT, using routine CBCT images. Our results show that different regions of the tumor change in different ways. Therefore, careful consideration should be taken when adapting RT.

GraphRegNet: Deep Graph Regularisation Networks on Sparse Keypoints for Dense Registration of 3D Lung CTs

In the last two years learning-based methods have started to show encouraging results in different supervised and unsupervised medical image registration tasks. Deep neural networks enable (near) real time applications through fast inference times and have tremendous potential for increased registration accuracies by task-specific learning. However, estimation of large 3D deformations, for example present in inhale to exhale lung CT or interpatient abdominal MRI registration, is still a major challenge for the widely adopted U-Net-like network architectures. Even when using multi-level strategies, current state-of-the-art DL registration results do not yet reach the high accuracy of conventional frameworks. To overcome the problem of large deformations for deep learning approaches, in this work, we present GraphRegNet, a sparse keypoint-based geometric network for dense deformable medical image registration. Similar to the successful 2D optical flow estimation of FlowNet or PWC-Net we leverage discrete dense displacement maps to facilitate the registration process. In order to cope with enormously increasing memory requirements when working with displacement maps in 3D medical volumes and to obtain a well-regularised and accurate deformation field we 1) formulate the registration task as the prediction of displacement vectors on a sparse irregular grid of distinctive keypoints and 2) introduce our efficient GraphRegNet for displacement regularisation, a combination of convolutional and graph neural network layers in a unified architecture. In our experiments on exhale to inhale lung CT registration we demonstrate substantial improvements (TRE below 1.4 mm) over other deep learning methods. Our code is publicly available at https://github.com/multimodallearning/graphregnet.

Ventilation measurements using fast-helical free-breathing computed tomography

PURPOSE

To examine the use of multiple fast-helical free breathing computed tomography (FHFBCT) scans for ventilation measurement.

METHODS

Ten patients were scanned 25 times in alternating directions using a FHFBCT protocol. Simultaneously, an abdominal pneumatic bellows was used as a real-time breathing surrogate. Regions-of-interest (ROIs) were selected from the upper right lungs of each patient for analysis. The ROIs were first registered using a published registration technique (pTV). A subsequent follow-up registration employed an objective function with two terms, a ventilation-adjusted Hounsfield Unit difference and a conservation-of-mass term labeled ΔΓ that denoted the difference between the deformation Jacobian and the tissue density ratio. The ventilations were calculated voxel-by-voxel as the slope of a first-order fit of the Jacobian as a function of the breathing amplitude.

RESULTS

The ventilations of the 10 patients showed different patterns and magnitudes. The average ventilation calculated from the deformation vector fields (DVFs) of the pTV and secondary registration was nearly identical, but the standard deviation of the voxel-to-voxel differences was approximately 0.1. The mean of the 90^th percentile values of ΔΓ was reduced from 0.153 to 0.079 between the pTV and secondary registration, implying first that the secondary registration improved the conservation-of-mass criterion by almost 50% and that on average the correspondence between the Jacobian and density ratios as demonstrated by ΔΓ was less than 0.1. This improvement occurred in spite of the average of the 90th percentile changes in the DVF magnitudes being only 0.58 mm.

CONCLUSIONS

This work introduces the use of multiple free-breathing CT scans for free-breathing ventilation measurements. The approach has some benefits over the traditional use of 4-dimensional CT (4DCT) or breath-hold scans. The benefit over 4DCT is that FHFBCT does not have sorting artifacts. The benefits over breath-hold scans include the relatively small motion induced by quiet respiration versus deep-inspiration breath hold and the potential for characterizing dynamic breathing processes that disappear during breath hold.

Disease activity in COPD: time to make imaging biomarkers a PET project?

FDG uptake on PET/CT is a potential biomarker of pulmonary inflammation in COPD and may reflect disease activity, but does it have the characteristics of a "good" biomarker? https://bit.ly/3AXheEZ.

Characterizing spatial differences between SPECT-ventilation and SPECT-perfusion in patients with lung cancer undergoing radiotherapy

This study investigates agreement between ventilation and perfusion for lung cancer patients undergoing radiotherapy. Ventilation-perfusion scans of nineteen patients with stage III lung cancer from a prospective protocol were compared using voxel-wise Spearman correlation-coefficients. The presented results show in about 25% of patients, ventilation and perfusion exhibit lower agreement.

Radiation-induced Hounsfield unit change correlates with dynamic CT perfusion better than 4DCT-based ventilation measures in a novel-swine model

To analyze radiation induced changes in Hounsfield units and determine their correlation with changes in perfusion and ventilation. Additionally, to compare the post-RT changes in human subjects to those measured in a swine model used to quantify perfusion changes and validate their use as a preclinical model. A cohort of 5 Wisconsin Miniature Swine (WMS) were studied. Additionally, 19 human subjects were recruited as part of an IRB approved clinical trial studying functional avoidance radiation therapy for lung cancer and were treated with SBRT. Imaging (a contrast enhanced dynamic perfusion CT in the swine and 4DCT in the humans) was performed prior to and post-RT. Jacobian elasticity maps were calculated on all 4DCT images. Contours were created from the isodose lines to discretize analysis into 10 Gy dose bins. B-spline deformable image registration allowed for voxel-by-voxel comparative analysis in these contours between timepoints. The WMS underwent a research course of 60 Gy in 5 fractions delivered locally to a target in the lung using an MRI-LINAC system. In the WMS subjects, the dose-bin contours were copied onto the contralateral lung, which received < 5 Gy for comparison. Changes in HU and changes in Jacobian were analyzed in these contours. Statistically significant (p < 0.05) changes in the mean HU value post-RT compared to pre-RT were observed in both the human and WMS groups at all timepoints analyzed. The HU increased linearly with dose for both groups. Strong linear correlation was observed between the changes seen in the swine and humans (Pearson coefficient > 0.97, p < 0.05) at all timepoints. Changes seen in the swine closely modeled the changes seen in the humans at 12 months post RT (slope = 0.95). Jacobian analysis showed between 30 and 60% of voxels were damaged post-RT. Perfusion analysis in the swine showed a statistically significant (p < 0.05) reduction in contrast inside the vasculature 3 months post-RT compared to pre-RT. The increases in contrast outside the vasculature was strongly correlated (Pearson Correlation 0.88) with the reduction in HU inside the vasculature but were not correlated with the changes in Jacobians. Radiation induces changes in pulmonary anatomy at 3 months post-RT, with a strong linear correlation with dose. The change in HU seen in the non-vessel lung parenchyma suggests this metric is a potential biomarker for change in perfusion. Finally, this work suggests that the WMS swine model is a promising pre-clinical model for analyzing radiation-induced changes in humans and poses several benefits over conventional swine models.

Geodesic density regression for correcting 4DCT pulmonary respiratory motion artifacts

Pulmonary respiratory motion artifacts are common in four-dimensional computed tomography (4DCT) of lungs and are caused by missing, duplicated, and misaligned image data. This paper presents a geodesic density regression (GDR) algorithm to correct motion artifacts in 4DCT by correcting artifacts in one breathing phase with artifact-free data from corresponding regions of other breathing phases. The GDR algorithm estimates an artifact-free lung template image and a smooth, dense, 4D (space plus time) vector field that deforms the template image to each breathing phase to produce an artifact-free 4DCT scan. Correspondences are estimated by accounting for the local tissue density change associated with air entering and leaving the lungs, and using binary artifact masks to exclude regions with artifacts from image regression. The artifact-free lung template image is generated by mapping the artifact-free regions of each phase volume to a common reference coordinate system using the estimated correspondences and then averaging. This procedure generates a fixed view of the lung with an improved signal-to-noise ratio. The GDR algorithm was evaluated and compared to a state-of-the-art geodesic intensity regression (GIR) algorithm using simulated CT time-series and 4DCT scans with clinically observed motion artifacts. The simulation shows that the GDR algorithm has achieved significantly more accurate Jacobian images and sharper template images, and is less sensitive to data dropout than the GIR algorithm. We also demonstrate that the GDR algorithm is more effective than the GIR algorithm for removing clinically observed motion artifacts in treatment planning 4DCT scans. Our code is freely available at https://github.com/Wei-Shao-Reg/GDR.

A generative adversarial network (GAN)-based technique for synthesizing realistic respiratory motion in the extended cardiac-torso (XCAT) phantoms

Objective. Synthesize realistic and controllable respiratory motions in the extended cardiac-torso (XCAT) phantoms by developing a generative adversarial network (GAN)-based deep learning technique.Methods. A motion generation model was developed using bicycle-GAN with a novel 4D generator. Input with the end-of-inhale (EOI) phase images and a Gaussian perturbation, the model generates inter-phase deformable-vector-fields (DVFs), which were composed and applied to the input to generate 4D images. The model was trained and validated using 71 4D-CT images from lung cancer patients and then applied to the XCAT EOI images to generate 4D-XCAT with realistic respiratory motions. A separate respiratory motion amplitude control model was built using decision tree regression to predict the input perturbation needed for a specific motion amplitude, and this model was developed using 300 4D-XCAT generated from 6 XCAT phantom sizes with 50 different perturbations for each size. In both patient and phantom studies, Dice coefficients for lungs and lung volume variation during respiration were compared between the simulated images and reference images. The generated DVF was evaluated by deformation energy. DVFs and ventilation maps of the simulated 4D-CT were compared with the reference 4D-CTs using cross correlation and Spearman's correlation. Comparison of DVFs and ventilation maps among the original 4D-XCAT, the generated 4D-XCAT, and reference patient 4D-CTs were made to show the improvement of motion realism by the model. The amplitude control error was calculated.Results. Comparing the simulated and reference 4D-CTs, the maximum deviation of lung volume during respiration was 5.8%, and the Dice coefficient reached at least 0.95 for lungs. The generated DVFs presented comparable deformation energy levels. The cross correlation of DVFs achieved 0.89 ± 0.10/0.86 ± 0.12/0.95 ± 0.04 along thex/y/zdirection in the testing group. The cross correlation of ventilation maps derived achieved 0.80 ± 0.05/0.67 ± 0.09/0.68 ± 0.13, and the Spearman's correlation achieved 0.70 ± 0.05/0, 60 ± 0.09/0.53 ± 0.01, respectively, in the training/validation/testing groups. The generated 4D-XCAT phantoms presented similar deformation energy as patient data while maintained the lung volumes of the original XCAT phantom (Dice = 0.95, maximum lung volume variation = 4%). The motion amplitude control models controlled the motion amplitude control error to be less than 0.5 mm.Conclusions. The results demonstrated the feasibility of synthesizing realistic controllable respiratory motion in the XCAT phantom using the proposed method. This crucial development enhances the value of XCAT phantoms for various 4D imaging and therapy studies.