Real‐time volumetric image reconstruction and 3D tumor localization based on a single x‐ray projection image for lung cancer radiotherapy

Li, Ruijiang; Jia, Xun; Lewis, John H.; Gu, Xuejun; Folkerts, Michael; Men, Chunhua; Jiang, Steve B

doi:10.1118/1.3426002

Cited by 111 publications

(145 citation statements)

References 18 publications

Supporting

Mentioning

142

Contrasting

Order By: Relevance

“…Besides the MM-FD method in this study, recently several other methods using similar deformation-based image estimation approaches have been developed, [45][46][47] which all have the potential to improve the dosimetric verification accuracy. Compared with these methods, the MM-FD method is proved more robust to the motion pattern/anatomical structure variations between 4D-CT and 4D-CBCT (Ref.…”

Section: D Comparison With Other Methodsmentioning

confidence: 99%

Dosimetric verification of lung cancer treatment using the CBCTs estimated from limited‐angle on‐board projections

2015

View full text Add to dashboard Cite

Purpose: Lung cancer treatment is susceptible to treatment errors caused by interfractional anatomical and respirational variations of the patient. On-board treatment dose verification is especially critical for the lung stereotactic body radiation therapy due to its high fractional dose. This study investigates the feasibility of using cone-beam (CB)CT images estimated by a motion modeling and free-form deformation (MM-FD) technique for on-board dose verification. Methods: Both digital and physical phantom studies were performed. Various interfractional variations featuring patient motion pattern change, tumor size change, and tumor average position change were simulated from planning CT to on-board images. The doses calculated on the planning CT (planned doses), the on-board CBCT estimated by MM-FD (MM-FD doses), and the on-board CBCT reconstructed by the conventional Feldkamp-Davis-Kress (FDK) algorithm (FDK doses) were compared to the on-board dose calculated on the "gold-standard" on-board images (gold-standard doses). The absolute deviations of minimum dose (∆D min ), maximum dose (∆D max ), and mean dose (∆D mean ), and the absolute deviations of prescription dose coverage (∆V 100% ) were evaluated for the planning target volume (PTV). In addition, 4D on-board treatment dose accumulations were performed using 4D-CBCT images estimated by MM-FD in the physical phantom study. The accumulated doses were compared to those measured using optically stimulated luminescence (OSL) detectors and radiochromic films. Results: Compared with the planned doses and the FDK doses, the MM-FD doses matched much better with the gold-standard doses. For the digital phantom study, the average (± standard deviation) ∆D min , ∆D max , ∆D mean , and ∆V 100% (values normalized by the prescription dose or the total PTV) between the planned and the gold-standard PTV doses were 32.9% (±28.6%), 3.0% (±2.9%), 3.8% (±4.0%), and 15.4% (±12.4%), respectively. The corresponding values of FDK PTV doses were 1.6% (±1.9%), 1.2% (±0.6%), 2.2% (±0.8%), and 17.4% (±15.3%), respectively. In contrast, the corresponding values of MM-FD PTV doses were 0.3% (±0.2%), 0.9% (±0.6%), 0.6% (±0.4%), and 1.0% (±0.8%), respectively. Similarly, for the physical phantom study, the average ∆D min , ∆D max , ∆D mean , and ∆V 100% of planned PTV doses were 38.1% (±30.8%), 3.5% (±5.1%), 3.0% (±2.6%), and 8.8% (±8.0%), respectively. The corresponding values of FDK PTV doses were 5.8% (±4.5%), 1.6% (±1.6%), 2.0% (±0.9%), and 9.3% (±10.5%), respectively. In contrast, the corresponding values of MM-FD PTV doses were 0.4% (±0.8%), 0.8% (±1.0%), 0.5% (±0.4%), and 0.8% (±0.8%), respectively. For the 4D dose accumulation study, the average (± standard deviation) absolute dose deviation (normalized by local doses) between the accumulated doses and the OSL measured doses was 3.3% (±2.7%). The average gamma index (3%/3 mm) between the accumulated doses and the radiochromic film measured doses was 94.5% (±2.5%). Conclusions: MM-FD estimated 4D-CBCT enables accurate on-board do...

show abstract

Section: D Comparison With Other Methodsmentioning

confidence: 99%

Dosimetric verification of lung cancer treatment using the CBCTs estimated from limited‐angle on‐board projections

2015

View full text Add to dashboard Cite

show abstract

“…In cooperation with multi-channel singular spectral analysis (MSSA), this can produce the future data based on the renewal knowledge from moment to moment. Lung motion model based on PCA has already been reported, which efficiently represents lung motion with few principal components [11]. On the other hand, these models have so far been used only for the image creation, and prediction has not been considered.…”

Section: Introductionmentioning

confidence: 99%

Dynamic Image Prediction Using Principal Component and Multi-Channel Singular Spectral Analysis: A Feasibility Study

Chhatkuli¹,

Demachi²,

Miyamoto³

et al. 2015

OJMI

View full text Add to dashboard Cite

Respiratory motion induces the limit in delivery accuracy due to the lack of the consideration of the anatomy motion in the treatment planning. Therefore, image-guided radiation therapy (IGRT) system plays an essential role in respiratory motion management and real-time tumor tracking in external beam radiation therapy. The objective of this research is the prediction of dynamic timeseries images considering the motion and the deformation of the tumor and to compensate the delay that occurs between the motion of the tumor and the beam delivery. For this, we propose a prediction algorithm for dynamic time-series images. Prediction is performed using principal component analysis (PCA) and multi-channel singular spectral analysis (MSSA). Using PCA, the motion can be denoted as a vector function and it can be estimated by its principal component which is the linear combination of eigen vectors corresponding to the largest eigen values. Timeseries set of 320-detector-row CT images from lung cancer patient and kilovolt (kV) fluoroscopic images from a moving phantom were used for the evaluation of the algorithm, and both image sets were successfully predicted by the proposed algorithm. The accuracy of prediction was quite high, more than 0.999 for CT images, whereas 0.995 for kV fluoroscopic images in cross-correlation coefficient value. This algorithm for image prediction makes it possible to predict the tumor images over the next breathing period with significant accuracy.

show abstract

“…Recent advances of the IGRT registration methods emphasize real-time computation and low-dose image acquisition. Russakoff et al [1,2], Khamene et al [3], Munbodh et al [4], Li et al [5,6] rejected the time-consuming 3D/3D registration and performed 2D/3D registration by optimizing similarity functions defined in the projection domain. Other than the optimization-based methods, Chou et al [7,8] recently introduced a faster and low-dose 2D/3D image registration by using a linear operator that approximates the deformation parameters.…”

Section: Introductionmentioning

confidence: 99%

Real-Time 2D/3D Deformable Registration Using Metric Learning

Chou

Pizer

2013

Medical Computer Vision. Recognition Techniques and Applications in Medical Imaging

View full text Add to dashboard Cite

Abstract. We present a novel 2D/3D deformable registration method, called Registration Efficiency and Accuracy through Learning Metric on Shape (REALMS ), that can support real-time Image-Guided Radiation Therapy (IGRT ). The method consists of two stages: planning-time learning and registration. In the planning-time learning, it firstly models the patient's 3D deformation space from the patient's time-varying 3D planning images using a low-dimensional parametrization. Secondly, it samples deformation parameters within the deformation space and generates corresponding simulated projection images from the deformed 3D image. Finally, it learns a Riemannian metric in the projection space for each deformation parameter. The learned distance metric forms a Gaussian kernel of a kernel regression that minimizes the leave-one-out regression residual of the corresponding deformation parameter. In the registration, REALMS interpolates the patient's 3D deformation parameters using the kernel regression with the learned distance metrics. Our test results showed that REALMS can localize the tumor in 10.89 ms (91.82 fps) with 2.56 ± 1.11 mm errors using a single projection image. These promising results show REALMS's high potential to support realtime, accurate, and low-dose IGRT.

show abstract

Real‐time volumetric image reconstruction and 3D tumor localization based on a single x‐ray projection image for lung cancer radiotherapy

Abstract: The authors have shown the feasibility of reconstructing volumetric images and localizing tumor positions in 3D in near real-time from a single x-ray image.

Cited by 111 publications

References 18 publications

Dosimetric verification of lung cancer treatment using the CBCTs estimated from limited‐angle on‐board projections

Dosimetric verification of lung cancer treatment using the CBCTs estimated from limited‐angle on‐board projections

Dynamic Image Prediction Using Principal Component and Multi-Channel Singular Spectral Analysis: A Feasibility Study

Real-Time 2D/3D Deformable Registration Using Metric Learning

Contact Info

Product

Resources

About