Real-Time Tumor Tracking During Radiotherapy

The efficacy of SBRT with focal boost is significantly limited by prostate/tumor motion and deformation during treatment. Many studies have shown that prostate motion is random, sporadic, and patient-specific. SBRT with focal boost requires steep dose gradients, and even small tumor motions could cause significant dose errors. The high fractional dose and low fraction number in SBRT magnify the impact of localization errors in a single fraction, leaving little or no room for a correction in the remaining fractions. Therefore, real-time prostate and tumor tracking is imperative for improving the outcome of prostate SBRT with focal boost.

Related publications:

Lei Y, Tian S, He X, Wang T, Wang B, Pretesh P, Jani A, Mao H, Curran W, Liu T and Yang X*. “Ultrasound Prostate Segmentation Based on Multi-Directional Deeply Supervised V-Net," Medical Physics, 46(6):3194-3206, 2019.

Zeng Q, Fu Y, Tian Z, Lei Y, Zhang Y, Wang T, Wang H, Mao H, Liu T, Curran W, Jani A, Patel P and Yang X*. “Label-Driven MRI-US Registration Using Weakly-Supervised Learning for MRI-guided Prostate Radiotherapy," Physics in Medicine and Biology, 65(13):135002, 2020.

Fu Y, Lei Y, Wang T, Patel P, Jani A, Mao H, Curran W, Liu T and Yang X*. "Biomechanically Constrained Non-rigid MR-TRUS Prostate Registration using Deep Learning based 3D Point Cloud Matching," Medical Image Analysis, 67,101845, 2021.

Fu Y, Wang T, Lei Y, Patel P, Jani A, Curran W, Liu T and Yang X*. “Deformable MR-CBCT Prostate Registration using Biomechanically Constrained Deep Learning Networks,” Medical Physics, 48(1):253-263, 2021.

Yang X*, Rossi P, Jani A, Mao H, Ogunleye T, Curran W and Liu T*. “A MR-TRUS Registration Method for Ultrasound-Guided Prostate Interventions.” Proc. of SPIE, 94151Y, 94151Y-9, 2015.

Lin Y, Liu T, Yang X, Wang W and Khan M. “Respiratory Induced Prostate Motion Using Wavelet Decomposition of the Real Time Electromagnetic Tracking Signal.” International Journal of Radiation Oncology • Biology • Physics (IJROBP), 87(2), 370-4, 2013.

Lin Y, Liu T, Wang W, Yang X and Khan M. “The Non-Gaussian Nature of Prostate Motion Based on Realtime Intra-fraction Tracking.” International Journal of Radiation Oncology • Biology • Physics (IJROBP), 87(2), 363-9, 2013.

During liver irradiation, it is paramount to accurately reproduce the position assumed by the patient during the CT simulation used to prepare the treatment plan. Liver motion leads to smearing of the dose distribution, causing reduced sharpness of the intended dose conformality. Therefore, there is an urgent need for a reliable real-time tracking method to ensure that the escalated dose is delivered to the tumor region as planned and to potentially reduce the treated volume of liver and Gastrointestinal structures.

Related publications:

Lei Y, Fu Y, Wang T, Liu Y, Higgins K, Curran W, Liu T and Yang X*. “4D-CT Deformable Image Registration Using Multiscale Unsupervised Deep Learning,” Physics in Medicine and Biology, 65(8), 085003, 2020.

Lei Y, Momin S, Roper J, Patel P, Curran WJ, Liu T and Yang X*. “Motion tracking in 3D ultrasound imaging based on Markov-like deep-learning-based deformable registration.” Proc. of SPIE, 11602:1160215, 2021.

Lei Y, Fu Y, Harms J, Wang T, Curran W, Liu T, Higgins K and Yang X*. 4D-CT Deformable Image Registration Using an Unsupervised Deep Convolutional Neural Network. In: Nguyen D., Xing L., Jiang S. (eds) Artificial Intelligence in Radiation Therapy. AIRT 2019. Lecture Notes in Computer Science, vol 11850. Springer, Cham.

Respiratory motion causes substantial anatomic changes, leading to significant dosimetric uncertainties in lung cancer radiotherapy. The impact of these uncertainties is amplified in lung stereotactic body radiation therapy (SBRT), which delivers very high, potent doses in relatively few fractions. Respiratory motion can vary between CT simulation, patient positioning and treatment delivery in its magnitude, baseline, period and regularity. Hence, in-treatment real-time on-board volumetric imaging is highly desired to provide the actual anatomy information during treatment delivery for in-treatment motion monitoring and management, and enable accurate marker-less tumor tracking by visualizing the time-resolved 3D anatomy change in real time during treatment delivery.

Related publications:

Lei Y, Zhen T, Wang T, Higgins K, Bradley J, Curran W, Liu T and Yang X*. “Deep learning-based Real-time Volumetric Imaging for Lung Stereotactic Body Radiation Therapy: A Proof-of-Concept Study,” Physics in Medicine and Biology, 65(23), 235003, 2020.

Fu Y, Lei Y, Wang T, Liu Y, Higgins K, Bradley J, Curran W, Liu T and Yang X*. “LungRegNet: An Unsupervised Deformable Image Registration Method for 4D-CT Lung,” Medical Physics, 2020. 47 (4):1763-1774, 2020. (Editor’s Choice)

Dong X, Lei Y, Wang T, Thomas M, Tang L, Curran W, Liu T and Yang X*. “Automatic Multi-Organ Segmentation in Thorax CT Images Using U-Net-GAN," Medical Physics, 46(5):2157-2168, 2019. (Editor’s Choice)