Evaluation of tumor localization in respiration motion-corrected cone-beam CT: prospective study in lung

Machine-learning-based prediction of the effectiveness of the delivered dose by exhale-gated radiotherapy for locally advanced lung cancer: The additional value of geometric over dosimetric parameters alone

Purpose

This study aimed to assess interfraction stability of the delivered dose distribution by exhale-gated volumetric modulated arc therapy (VMAT) or intensity-modulated arc therapy (IMAT) for lung cancer and to determine dominant prognostic dosimetric and geometric factors.

Methods

Clinical target volume (CTV_Plan) from the planning CT was deformed to the exhale-gated daily CBCT scans to determine CTV_i, treated by the respective dose fraction. The equivalent uniform dose of the CTV_i was determined by the power law (gEUD_i) and cell survival model (EUD_iSF) as effectiveness measure for the delivered dose distribution. The following prognostic factors were analyzed: (I) minimum dose within the CTV_i (D_{min_i}), (II) Hausdorff distance (HDD_i) between CTV_i and CTV_Plan, (III) doses and deformations at the point in CTV_Plan at which the global minimum dose over all fractions per patient occurs (PD_{min_global_i}), and (IV) deformations at the point over all CTV_i margins per patient with the largest Hausdorff distance (HDPw_orst). Prognostic value and generalizability of the prognostic factors were examined using cross-validated random forest or multilayer perceptron neural network (MLP) classifiers. Dose accumulation was performed using back deformation of the dose distribution from CTV_i to CTV_Plan.

Results

Altogether, 218 dose fractions (10 patients) were evaluated. There was a significant interpatient heterogeneity between the distributions of the normalized gEUD_i values (p<0.0001, Kruskal-Wallis tests). Accumulated gEUD over all fractions per patient was 1.004-1.023 times of the prescribed dose. Accumulation led to tolerance of ~20% of fractions with gEUD_i <93% of the prescribed dose. Normalized D_min >60% was associated with predicted gEUD values above 95%. D_min had the highest importance for predicting the gEUD over all analyzed prognostic parameters by out-of-bag loss reduction using the random forest procedure. Cross-validated random forest classifier based on D_min as the sole input had the largest Pearson correlation coefficient (R=0.897) in comparison to classifiers using additional input variables. The neural network performed better than the random forest classifier, and the gEUD values predicted by the MLP classifier with D_min as the sole input were correlated with the gEUD values characterized by R=0.933 (95% CI, 0.913-0.948). The performance of the full MLP model with all geometric input parameters was slightly better (R=0.952) than that based on D_min (p=0.0034, Z-test).

Conclusion

Accumulated dose distributions over the treatment series were robust against interfraction CTV deformations using exhale gating and online image guidance. D_min was the most important parameter for gEUD prediction for a single fraction. All other parameters did not lead to a markedly improved generalizable prediction. Dosimetric information, especially location and value of D_min within the CTV _i , are vital information for image-guided radiation treatment.

A modern review of the uncertainties in volumetric imaging of respiratory-induced target motion in lung radiotherapy

Radiotherapy has become a critical component for the treatment of all stages and types of lung cancer, often times being the primary gateway to a cure. However, given that radiation can cause harmful side effects depending on how much surrounding healthy tissue is exposed, treatment of the lung can be particularly challenging due to the presence of moving targets. Careful implementation of every step in the radiotherapy process is absolutely integral for attaining optimal clinical outcomes. With the advent and now widespread use of stereotactic body radiation therapy (SBRT), where extremely large doses are delivered, accurate, and precise dose targeting is especially vital to achieve an optimal risk to benefit ratio. This has largely become possible due to the rapid development of image-guided technology. Although imaging is critical to the success of radiotherapy, it can often be plagued with uncertainties due to respiratory-induced target motion. There has and continues to be an immense research effort aimed at acknowledging and addressing these uncertainties to further our abilities to more precisely target radiation treatment. Thus, the goal of this article is to provide a detailed review of the prevailing uncertainties that remain to be investigated across the different imaging modalities, as well as to highlight the more modern solutions to imaging motion and their role in addressing the current challenges.

Technical Note: 3D localization of lung tumors on cone beam CT projections via a convolutional recurrent neural network

PURPOSE

To design a convolutional recurrent neural network (CRNN) that calculates three-dimensional (3D) positions of lung tumors from continuously acquired cone beam computed tomography (CBCT) projections, and facilitates the sorting and reconstruction of 4D-CBCT images.

METHOD

Under an IRB-approved clinical lung protocol, kilovoltage (kV) projections of the setup CBCT were collected in free-breathing. Concurrently, an electromagnetic signal-guided system recorded motion traces of three transponders implanted in or near the tumor. Convolutional recurrent neural network was designed to utilize a convolutional neural network (CNN) for extracting relevant features of the kV projections around the tumor, followed by a recurrent neural network for analyzing the temporal patterns of the moving features. Convolutional recurrent neural network was trained on the simultaneously collected kV projections and motion traces, subsequently utilized to calculate motion traces solely based on the continuous feed of kV projections. To enhance performance, CRNN was also facilitated by frequent calibrations (e.g., at 10° gantry rotation intervals) derived from cross-correlation-based registrations between kV projections and templates created from the planning 4DCT. Convolutional recurrent neural network was validated on a leave-one-out strategy using data from 11 lung patients, including 5500 kV images. The root-mean-square error between the CRNN and motion traces was calculated to evaluate the localization accuracy.

RESULT

Three-dimensional displacement around the simulation position shown in the Calypso traces was 3.4 ± 1.7 mm. Using motion traces as ground truth, the 3D localization error of CRNN with calibrations was 1.3 ± 1.4 mm. CRNN had a success rate of 86 ± 8% in determining whether the motion was within a 3D displacement window of 2 mm. The latency was 20 ms when CRNN ran on a high-performance computer cluster.

CONCLUSIONS

CRNN is able to provide accurate localization of lung tumors with aid from frequent recalibrations using the conventional cross-correlation-based registration approach, and has the potential to remove reliance on the implanted fiducials.

Impact of the Time Proportion of Respiratory Phases on Dosimetry in SBRT of Lung Tumor Near the Chest Wall or Diaphragm

Purpose:

Large tumor motion often leads to larger dosimetric variation, especially in lung tumors located in lower lobe and adhered to chest wall or diaphragm. The purpose of this work is to discuss the impact of the time proportion of each respiratory phase on dosimetry in stereotactic body radiation therapy with lung cancer tumor close to chest wall or diaphragm.

Methods:

Participants include 14 patients with lung cancer located in the lower lobe. Each patient received treatment planning 3-dimensional computed tomography and an additional 4-dimensional computed tomography simulation under free-breathing condition. The percentage of time intervals for each respiratory phase in a whole respiratory cycle was calculated from respiratory motion curves recorded during 4-dimensional computed tomography scanning. Treatment plan was made upon treatment planning 3-dimensional computed tomography and then transformed onto each image of 4-dimensional computed tomography. The transformed doses on each image of 4-dimensional computed tomography were accumulated with equal weight or with weight of time proportion for each respiratory phase.

Results:

Compared to treatment planning 3-dimensional computed tomography dose, the mean dose of tumor, affected lung, contralateral lung, bilateral lungs, and V₂₀ of affected lung decreased by 2.7%, 4.5%, 1.5%, 1.2%, and 4.1%, respectively, after equal-weighted accumulation, while mean dose of heart increased by 3.6% (P < .05). Accumulated dose with weight of actual time proportion decreased in the mean dose of tumor, affected lung, contralateral lung, bilateral lungs, and V₂₀ of affected lung by 2.37%, 5.19%, 3.61%, 3.46%, and 5.08%, respectively compared to treatment planning 3-dimensional computed tomography dose, but mean dose of heart increased by 5.12% (P < .05).

Conclusions:

Doses received by tumor, lungs, and heart changed more significantly after time-weighted 4-dimensional accumulation than equal-weight 4-dimensional accumulation. Utilizing 4-dimensional computed tomography and deformable image registration to introduce time proportions of each respiratory phase to dose distribution evaluation is of significance for accuracy in lung cancer during stereotactic body radiation therapy treatment.

Design and validation of a MV/kV imaging-based markerless tracking system for assessing real-time lung tumor motion

PURPOSE

Localizing lung tumors during treatment delivery is critical for managing respiratory motion, ensuring tumor coverage, and reducing toxicities. The purpose of this project is to develop a real-time system that performs markerless tracking of lung tumors using simultaneously acquired MV and kV images during radiotherapy of lung cancer with volumetric modulated arc therapy.

METHOD

Continuous MV/kV images were simultaneously acquired during dose delivery. In the subsequent analysis, a gantry angle-specific region of interest was defined according to the treatment aperture. After removing imaging artifacts, processed MV/kV images were directly registered to the corresponding daily setup cone-beam CT (CBCT) projections that served as reference images. The registration objective function consisted of a sum of normalized cross-correlation, weighted by the contrast-to-noise ratio of each MV and kV image. The calculated 3D shifts of the tumor were corrected by the displacements between the CBCT projections and the planning respiratory correlated CT (RCCT) to generate motion traces referred to a specific respiratory phase. The accuracy of the algorithm was evaluated on both anthropomorphic phantom and patient studies. The phantom consisted of localizing a 3D printed tumor, embedded in a thorax phantom, in an arc delivery. In an IRB-approved study, data were obtained from VMAT treatments of two lung cancer patients with three electromagnetic (Calypso) beacon transponders implanted in airways near the lung tumor.

RESULT

In the phantom study, the root mean square error (RMSE) between the registered and actual (programmed couch movement) target position was 1.2 mm measured by the MV/kV imaging system, which was smaller compared to the MV or kV alone, of 4.1 and 1.3 mm, respectively. In the patient study, the mean and standard deviation discrepancy between electromagnetic-based tumor position and the MV/KV-markerless approach was -0.2 ± 0.6 mm, 0.2 ± 1.0 mm, and -1.2 ± 1.5 mm along the superior-inferior, anterior-posterior, and left-right directions, respectively; resulting in a 3D displacement discrepancy of 2.0 ± 1.1 mm. Poor contrast around the tumor was the main contribution to registration uncertainties.

CONCLUSION

The combined MV/kV imaging system can provide real-time 3D localization of lung tumor, with comparable accuracy to the electromagnetic-based system when features of tumors are detectable. Careful design of a registration algorithm and a VMAT plan that maximizes the tumor visibility are key elements for a successful MV/KV localization strategy.

Data-driven respiratory motion compensation for four-dimensional cone-beam computed tomography (4D-CBCT) using groupwise deformable registration

PURPOSE

To demonstrate the feasibility of using a purely data-driven, a posteriori respiratory motion modeling and reconstruction compensation method to improve 4D-CBCT image quality under clinically relevant image acquisition conditions.

METHODS

Evaluated workflows that utilized a combination of groupwise deformable image registration and motion-compensated image reconstruction algorithms. Groupwise registration is an approach that simultaneously registers all temporal frames of a 4D image to a common reference instead of one at a time so as to minimize the influence of any individual time point on the global smoothness or accuracy of the resulting deformation model. Four-dimensional cone-beam CT (4D-CBCT) Feldkamp-Davis-Kress (FDK) reconstructions were registered to either iteratively computed mean respiratory phase (mean-frame) or preselected respiratory phase (fixed-frame) reference images to model respiratory motion. The resulting 4D transformations were used to deform projection data during the FDK backprojection operation to create motion-compensated reconstructions. Tissue interface sharpness (TIS) was defined as the slope of a sigmoid curve fit to a mobile tissue boundary and was used to evaluate image quality in regions susceptible to motion artifacts. Image quality improvement was assessed for 19 clinical cases by evaluating mitigation of view aliasing artifacts, TIS, image noise reduction, and contrast for implanted fiducial markers.

RESULTS

Average (standard deviation) diaphragm TIS recovery relative to initial 4D-CBCT reconstructions was observed to be 87% (46%) using fixed-frame registration alone; 87% (47%) using fixed frame with motion-compensated reconstruction; 101% (68%) using mean-frame registration alone; and 99% (65%) using mean frame with motion-compensated reconstruction. Noise was reduced in sampled soft tissue ROIs by 58% for both fixed-frame registration and registration with motion compensation and by 57% and 58% on average for the corresponding mean-frame methods, respectively. Average improvement in local CNR was observed to be respectively 93% and 98% for fixed-frame registration and registration with motion compensation methods and 116% and 111% for the corresponding mean-frame methods.

CONCLUSION

Data-driven groupwise registration and motion-compensated reconstruction offer a feasible means of improving the quality of 4D-CBCT images acquired under clinical conditions. The addition of motion compensation reconstruction after groupwise registration visibly reduced the impact of view aliasing artifacts for the clinical image datasets studied.

Hitting a Moving Target: Successful Management of Diffuse Large B-cell Lymphoma Involving the Mesentery With Volumetric Image-guided Intensity Modulated Radiation Therapy

INTRODUCTION

We report successful treatment of mesenteric diffuse large B-cell lymphoma (DLBCL) using localized involved site radiation therapy (ISRT), intensity modulated radiation therapy (IMRT), and daily computed tomography (CT)-image guidance.

PATIENTS AND METHODS

Patients with mesenteric DLBCL treated with RT between 2011 and 2017 were reviewed. Clinical and treatment characteristics were analyzed for an association with local control, progression-free survival (PFS), and overall survival.

RESULTS

Twenty-three patients were eligible. At diagnosis, the median age was 52 years (range, 38-76 years), and 57% (n = 13) had stage I/II DLBCL. All patients received frontline chemotherapy (ChT) (R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone], n = 19; dose-adjusted R-EPOCH [rituximab, etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin], n = 4) with median 6 cycles. Prior to RT, salvage ChT for refractory DLBCL was given to 43% (n = 10) and autologous stem cell transplantation was administered in 13% (n = 3). At the time of RT, positron emission tomography-CT revealed 5-point scale of 1 to 3 (48%; n = 11), 4 (9%; n = 2), and 5 (44%; n = 10). All patients received IMRT, daily CT imaging, and ISRT. The median RT dose was 40 Gy (range, 16.2-49.4 Gy). Relapse or progression occurred in 22% (n = 5). At a median follow-up of 37 months, the 3-year local control, PFS, and overall survival rates were 80%, 75%, and 96%, respectively. Among patients treated with RT after complete metabolic response to frontline ChT (n = 8), the 3-year PFS was 100%, compared with 61% for patients with a history of chemorefractory DLBCL (n = 15; P = .055). Four of the 5 relapses occurred in patients with 5-point scale of 5 prior to RT (P = .127).

CONCLUSION

Mesenteric involvement of DLBCL can be successfully targeted with localized ISRT fields using IMRT and daily CT-image guidance.

Evaluation of respiratory motion-corrected cone-beam CT at end expiration in abdominal radiotherapy sites: a prospective study

BACKGROUND

Cone beam computed tomography (CBCT) for radiotherapy image guidance suffers from respiratory motion artifacts. This limits soft tissue visualization and localization accuracy, particularly in abdominal sites. We report on a prospective study of respiratory motion-corrected (RMC)-CBCT to evaluate its efficacy in localizing abdominal organs and improving soft tissue visibility at end expiration.

MATERIAL AND METHODS

In an IRB approved study, 11 patients with gastroesophageal junction (GEJ) cancer and five with pancreatic cancer underwent a respiration-correlated CT (4DCT), a respiration-gated CBCT (G-CBCT) near end expiration and a one-minute free-breathing CBCT scan on a single treatment day. Respiration was recorded with an external monitor. An RMC-CBCT and an uncorrected CBCT (NC-CBCT) were computed from the free-breathing scan, based on a respiratory model of deformations derived from the 4DCT. Localization discrepancy was computed as the 3D displacement of the GEJ region (GEJ patients), or gross tumor volume (GTV) and kidneys (pancreas patients) in the NC-CBCT and RMC-CBCT relative to their positions in the G-CBCT. Similarity of soft-tissue features was measured using a normalized cross correlation (NCC) function.

RESULTS

Localization discrepancy from the end-expiration G-CBCT was reduced for RMC-CBCT compared to NC-CBCT in eight of eleven GEJ cases (mean ± standard deviation, respectively, 0.21 ± 0.11 and 0.43 ± 0.28 cm), in all five pancreatic GTVs (0.26 ± 0.21 and 0.42 ± 0.29 cm) and all ten kidneys (0.19 ± 0.13 and 0.51 ± 0.25 cm). Soft-tissue feature similarity around GEJ was higher with RMC-CBCT in nine of eleven cases (NCC =0.48 ± 0.20 and 0.43 ± 0.21), and eight of ten kidneys (0.44 ± 0.16 and 0.40 ± 0.17).

CONCLUSIONS

In a prospective study of motion-corrected CBCT in GEJ and pancreas, RMC-CBCT yielded improved organ visibility and localization accuracy for gated treatment at end expiration in the majority of cases.

Analysis of the Setup Uncertainty and Margin of the Daily ExacTrac 6D Image Guide System for Patients with Brain Tumors

This study evaluated the setup uncertainties for brain sites when using BrainLAB's ExacTrac X-ray 6D system for daily pretreatment to determine the optimal planning target volume (PTV) margin. Between August 2012 and April 2015, 28 patients with brain tumors were treated by daily image-guided radiotherapy using the BrainLAB ExacTrac 6D image guidance system of the Novalis-Tx linear accelerator. DUONTM (Orfit Industries, Wijnegem, Belgium) masks were used to fix the head. The radiotherapy was fractionated into 27-33 treatments. In total, 844 image verifications were performed for 28 patients and used for the analysis. The setup corrections along with the systematic and random errors were analyzed for six degrees of freedom in the translational (lateral, longitudinal, and vertical) and rotational (pitch, roll, and yaw) dimensions. Optimal PTV margins were calculated based on van Herk et al.'s [margin recipe = 2.5∑ + 0.7σ - 3 mm] and Stroom et al.'s [margin recipe = 2∑ + 0.7σ] formulas. The systematic errors (∑) were 0.72, 1.57, and 0.97 mm in the lateral, longitudinal, and vertical translational dimensions, respectively, and 0.72°, 0.87°, and 0.83° in the pitch, roll, and yaw rotational dimensions, respectively. The random errors (σ) were 0.31, 0.46, and 0.54 mm in the lateral, longitudinal, and vertical rotational dimensions, respectively, and 0.28°, 0.24°, and 0.31° in the pitch, roll, and yaw rotational dimensions, respectively. According to van Herk et al.'s and Stroom et al.'s recipes, the recommended lateral PTV margins were 0.97 and 1.66 mm, respectively; the longitudinal margins were 1.26 and 3.47 mm, respectively; and the vertical margins were 0.21 and 2.31 mm, respectively. Therefore, daily setup verifications using the BrainLAB ExacTrac 6D image guide system are very useful for evaluating the setup uncertainties and determining the setup margin.

Accuracy of image guidance using free-breathing cone-beam computed tomography for stereotactic lung radiotherapy

Movement of the target object during cone-beam computed tomography (CBCT) leads to motion blurring artifacts. The accuracy of manual image matching in image-guided radiotherapy depends on the image quality. We aimed to assess the accuracy of target position localization using free-breathing CBCT during stereotactic lung radiotherapy. The Vero4DRT linear accelerator device was used for the examinations. Reference point discrepancies between the MV X-ray beam and the CBCT system were calculated using a phantom device with a centrally mounted steel ball. The precision of manual image matching between the CBCT and the averaged intensity (AI) images restructured from four-dimensional CT (4DCT) was estimated with a respiratory motion phantom, as determined in evaluations by five independent operators. Reference point discrepancies between the MV X-ray beam and the CBCT image-guidance systems, categorized as left-right (LR), anterior-posterior (AP), and superior-inferior (SI), were 0.33 ± 0.09, 0.16 ± 0.07, and 0.05 ± 0.04 mm, respectively. The LR, AP, and SI values for residual errors from manual image matching were -0.03 ± 0.22, 0.07 ± 0.25, and -0.79 ± 0.68 mm, respectively. The accuracy of target position localization using the Vero4DRT system in our center was 1.07 ± 1.23 mm (2 SD). This study experimentally demonstrated the sufficient level of geometric accuracy using the free-breathing CBCT and the image-guidance system mounted on the Vero4DRT. However, the inter-observer variation and systematic localization error of image matching substantially affected the overall geometric accuracy. Therefore, when using the free-breathing CBCT images, careful consideration of image matching is especially important.

Four dimensional digital tomosynthesis using on-board imager for the verification of respiratory motion

Purpose

To evaluate respiratory motion of a patient by generating four-dimensional digital tomosynthesis (4D DTS), extracting respiratory signal from patients' on-board projection data, and ensuring the feasibility of 4D DTS as a localization tool for the targets which have respiratory movement.

Methods and Materials

Four patients with lung and liver cancer were included to verify the feasibility of 4D-DTS with an on-board imager. CBCT acquisition (650–670 projections) was used to reconstruct 4D DTS images and the breath signal of the patients was generated by extracting the motion of diaphragm during data acquisition. Based on the extracted signal, the projection data was divided into four phases: peak-exhale phase, mid-inhale phase, peak-inhale phase, and mid-exhale phase. The binned projection data was then used to generate 4D DTS, where the total scan angle was assigned as ±22.5° from rotation center, centered on 0° and 180° for coronal “half-fan” 4D DTS, and 90° and 270° for sagittal “half-fan” 4D DTS. The result was then compared with 4D CBCT which we have also generated with the same phase distribution.

Results

The motion of the diaphragm was evident from the 4D DTS results for peak-exhale, mid-inhale, peak-inhale and mid-exhale phase assignment which was absent in 3D DTS. Compared to the result of 4D CBCT, the view aliasing effect due to arbitrary angle reconstruction was less severe. In addition, the severity of metal artifacts, the image distortion due to presence of metal, was less than that of the 4D CBCT results.

Conclusion

We have implemented on-board 4D DTS on patients data to visualize the movement of anatomy due to respiratory motion. The results indicate that 4D-DTS could be a promising alternative to 4D CBCT for acquiring the respiratory motion of internal organs just prior to radiotherapy treatment.