A back propagation neural network based respiratory motion modelling method.

BACKGROUND: This study presents the development of a backpropagation neural network-based respiratory motion modelling method (BP-RMM) for precisely tracking arbitrary points within lung tissue throughout free respiration, encompassing deep inspiration and expiration phases. METHODS: Internal and external respiratory data from four-dimensional computed tomography (4DCT) are processed using various artificial intelligence algorithms. Data augmentation through polynomial interpolation is employed to enhance dataset robustness. A BP neural network is then constructed to comprehensively track lung tissue movement. RESULTS: The BP-RMM demonstrates promising accuracy. In cases from the public 4DCT dataset, the average target registration error (TRE) between authentic deep respiration phases and those forecasted by BP-RMM for 75 marked points is 1.819 mm. Notably, TRE for normal respiration phases is significantly lower, with a minimum error of 0.511 mm. CONCLUSIONS: The proposed method is validated for its high accuracy and robustness, establishing it as a promising tool for surgical navigation within the lung.

Deep match: A zero-shot framework for improved fiducial-free respiratory motion tracking.

BACKGROUND AND PURPOSE: Motion management is essential to reduce normal tissue exposure and maintain adequate tumor dose in lung stereotactic body radiation therapy (SBRT). Lung SBRT using an articulated robotic arm allows dynamic tracking during radiation dose delivery. Two stereoscopic X-ray tracking modes are available - fiducial-based and fiducial-free tracking. Although X-ray detection of implanted fiducials is robust, the implantation procedure is invasive and inapplicable to some patients and tumor locations. Fiducial-free tracking relies on tumor contrast, which challenges the existing tracking algorithms for small (e.g., <15 mm) and/or tumors obscured by overlapping anatomies. To markedly improve the performance of fiducial-free tracking, we proposed a deep learning-based template matching algorithm - Deep Match. METHOD: Deep Match consists of four self-definable stages - training-free feature extractor, similarity measurements for location proposal, local refinements, and uncertainty level prediction for constructing a more trustworthy and versatile pipeline. Deep Match was validated on a 10 (38 fractions; 2661 images) patient cohort whose lung tumor was trackable on one X-ray view, while the second view did not offer sufficient conspicuity for tumor tracking using existing methods. The patient cohort was stratified into subgroups based on tumor sizes (<10 mm, 10-15 mm, and >15 mm) and tumor locations (with/without thoracic anatomy overlapping). RESULTS: On X-ray views that conventional methods failed to track the lung tumor, Deep Match achieved robust performance as evidenced by >80 % 3 mm-Hit (detection within 3 mm superior/inferior margin from ground truth) for 70 % of patients and <3 mm superior/inferior distance (SID) ∼1 mm standard deviation for all the patients. CONCLUSION: Deep Match is a zero-shot learning network that explores the intrinsic neural network benefits without training on patient data. With Deep Match, fiducial-free tracking can be extended to more patients with small tumors and with tumors obscured by overlapping anatomy.

Analysis of the influence of tracking error of Xsight lung tracking system caused by cardiac beating / 中华放射肿瘤学杂志

Objective:To analyze the influence of tracking error of Xsight lung tracking system caused by cardiac beating.Methods:48 patients with lung tumors adjacent to the heart were enrolled into this study. The tumor movement curves were collected by the Xsight lung tracking system and recorded in the treatment log files during the Cyberknife treatment process. The curves were subject to filtering analysis and the respiratory motion of < 1 Hz and the cardiac beating motion of > 1 Hz were separated. According to the filtering results, the patient treatment tracking data were divided into two groups based on whether the cardiac beating wave of >1 Hz existed. The tracking errors were statistically compared between two groups based on the X-ray imaging data collected by Xsight lung tracking system during treatment.Results:For the fractionation with cardiac beat information, the tracking errors of the patient′s related models were (1.45 ± 0.99), (0.46 ± 0.21) and (0.70 ± 0.54) mm in the left-right, superior-inferior and anterior-posterior direction, respectively. For the fractionation without cardiac beat information, the tracking errors of the patient′s related models were (1.52 ± 1.17), (0.63 ± 0.37) and (1.07 ± 0.62) mm in the left-right, superior-inferior and anterior-posterior direction, respectively. The tracking errors in the superior-inferior and anterior-posterior direction of patients with accurate cardiac beat models were 28.34% and 34.86% less than those of their counterparts without accurate cardiac beat models and there was significant difference (both P<0.05). Conclusion:The tracking accuracy of Xsight lung tracking system will be significantly improved if the cardiac beat model is accurately established.

Verification of X-sight Lung Tracking System in the CyberKnife / 의학물리

To track moving tumor in real time, CyberKnife system imports a technique of the synchrony respiratory tracking system. The fiducial marker which are detectable in X-ray images were demand in CyberKnife Robotic radiosurgery system. It issued as reference markers to locate and track tumor location during patient alignment and treatment delivery. Fiducial marker implantation is an invasive surgical operation that carries a relatively high risk of pneumothorax. Most recently, it was developed a direct lung tumor registration method that does not require the use of fiducials. The purpose of this study is to measure the accuracy of target applying X-sight lung tracking using the Gafchromic film in dynamic moving thorax phantom. The X-sight Lung Tracking quality assurance motion phantom simulates simple respiratory motion of a lung tumor and provides Gafchromic dosimetry film-based test capability at locations inside the phantom corresponding to a typical lung tumor. The total average error for the X-sight Lung Tracking System with a moving target was 0.85+/-0.22 mm. The results were considered reliable and applicable for lung tumor treatment in CyberKnife radiosurgery system. Clinically, breathing patterns of patients may vary during radiation therapy. Therefore, additional studies with a set real patient data are necessary to evaluate the target accuracy for the X-sight Lung Tracking system.

Verification of X-sight Lung Tracking System in the CyberKnife / 의학물리

To track moving tumor in real time, CyberKnife system imports a technique of the synchrony respiratory tracking system. The fiducial marker which are detectable in X-ray images were demand in CyberKnife Robotic radiosurgery system. It issued as reference markers to locate and track tumor location during patient alignment and treatment delivery. Fiducial marker implantation is an invasive surgical operation that carries a relatively high risk of pneumothorax. Most recently, it was developed a direct lung tumor registration method that does not require the use of fiducials. The purpose of this study is to measure the accuracy of target applying X-sight lung tracking using the Gafchromic film in dynamic moving thorax phantom. The X-sight Lung Tracking quality assurance motion phantom simulates simple respiratory motion of a lung tumor and provides Gafchromic dosimetry film-based test capability at locations inside the phantom corresponding to a typical lung tumor. The total average error for the X-sight Lung Tracking System with a moving target was 0.85+/-0.22 mm. The results were considered reliable and applicable for lung tumor treatment in CyberKnife radiosurgery system. Clinically, breathing patterns of patients may vary during radiation therapy. Therefore, additional studies with a set real patient data are necessary to evaluate the target accuracy for the X-sight Lung Tracking system.