Tumor Tracking Method Research Articles

Unlocking Overexpressed Membrane Proteins to Guide Breast Cancer Precision Medicine.

Breast cancer (BC) is a prevalent form of cancer affecting women worldwide. However, the effectiveness of current BC drugs is limited by issues such as systemic toxicity, drug resistance, and severe side effects. Consequently, there is an urgent need for new therapeutic targets and improved tumor tracking methods. This study aims to address these challenges by proposing a strategy for identifying membrane proteins in tumors that can be targeted for specific BC therapy and diagnosis. The strategy involves the analyses of gene expressions in breast tumor and non-tumor tissues and other healthy tissues by using comprehensive bioinformatics analysis from The Cancer Genome Atlas (TCGA), UALCAN, TNM Plot, and LinkedOmics. By employing this strategy, we identified four transcripts (LRRC15, EFNA3, TSPAN13, and CA12) that encoded membrane proteins with an increased expression in BC tissue compared to healthy tissue. These four transcripts also demonstrated high accuracy, specificity, and accuracy in identifying tumor samples, as confirmed by the ROC curve. Additionally, tissue microarray (TMA) analysis revealed increased expressions of the four proteins in tumor tissues across all molecular subtypes compared to the adjacent breast tissue. Moreover, the analysis of human interactome data demonstrated the important roles of these proteins in various cancer-related pathways. Taken together, these findings suggest that LRRC15, EFNA3, TSPAN13, and CA12 can serve as potential biomarkers for improving cancer diagnosis screening and as suitable targets for therapy with reduced side effects and enhanced efficacy.

Feasibility study of deep learning-based markerless real-time lung tumor tracking with orthogonal X-ray projection images.

The feasibility of a deep learning-based markerless real-time tumor tracking (RTTT) method was retrospectively studied with orthogonal kV X-ray images and clinical tracking records acquired during lung cancer treatment. Ten patients with lung cancer treated with marker-implanted RTTT were included. The prescription dose was 50Gy in four fractions, using seven- to nine-port non-coplanar static beams. This corresponds to 14-18 X-ray tube angles for an orthogonal X-ray imaging system rotating with the gantry. All patients underwent 10 respiratory phases four-dimensional computed tomography. After a data augmentation approach, for each X-ray tube angle of a patient, 2250 digitally reconstructed radiograph (DRR) images with gross tumor volume (GTV) contour labeled were obtained. These images were adopted to train the patient and X-ray tube angle-specific GTV contour prediction model. During the testing, the model trained with DRR images predicted GTV contour on X-ray projection images acquired during treatment. The predicted three-dimensional (3D) positions of the GTV were calculated based on the centroids of the contours in the orthogonal images. The 3D positions of GTV determined by the marker-implanted RTTT during the treatment were considered as the ground truth. The 3D deviations between the prediction and the ground truth were calculated to evaluate the performance of the model. The median GTV volume and motion range were 7.42 (range, 1.18-25.74) cm3 and 22 (range, 11-28) mm, respectively. In total, 8993 3D position comparisons were included. The mean calculation time was 85ms per image. The overall median value of the 3D deviation was 2.27 (interquartile range: 1.66-2.95) mm. The probability of the 3D deviation smaller than 5mm was 93.6%. The evaluation results and calculation efficiency show the proposed deep learning-based markerless RTTT method may be feasible for patients with lung cancer.

A tumor tracking method in an ultrasound-guided radiofrequency ablation support system using deep learning

A tumor tracking method in an ultrasound-guided radiofrequency ablation support system using deep learning

Performance of Makerless Tracking for Gimbaled Dynamic Tumor Tracking

Background and PurposeThe purpose of this work is to report the workflow and the accuracy of the new markerless dynamic tumor tracking (MLDTT) method of the Vero 4DRT system introduced with ExacTrac 3.6.1. Material and MethodsPhantom measurements were performed to assess the accuracy of the MLDTT algorithm by using the QA-tool which is provided by the vendor. A patient breathing curve was used as the motion trajectory of the phantom and the target positions detected by the MLDTT algorithm were compared to the defined positions. Furthermore, eight patients have been treated with MLDTT between May 2018 and July 2019. A log-file analysis is used to evaluate MLDTT treatment data. ResultsThe accuracy of the MLDTT detection is 0.12mm ± 0.12mm, 0.12mm ± 0.11mm, 0.20mm ± 0.21mm for the x-, y-, z-direction, respectively. These values are comparable to the accuracy of marker based DTT at the Vero system. The median treatment time was 21min 34seconds and 175kV images were acquired during treatment for monitoring the target motion. ConclusionThe accuracy of the MLDTT algorithm is comparable to the marker based approach and the accuracy reported for the XSight Lung of the CyberKnife. Eight patients were treated successfully using MLDTT and the treatment times are comparable to a standard DTT treatment.

A Novel Markerless Lung Tumor-Tracking Method Using Treatment MV Beam Imaging

A novel method was developed to track lung tumor motion in real time during radiation therapy with the purpose to allow target radiation dose escalation while simultaneously reducing the dose to sensitive structures, thereby increasing local control without increasing toxicity. This method analyzes beam’s eye view radiation therapy treatment megavoltage (MV) images with simulated digitally reconstructed radiographs (DRRs) as references. Instead of comparing global DRRs with projection images, this method incorporates a technique that divides the global composite DRR and the corresponding MV projection into sub-images called tiles. Registration is performed independently on tile pairs in order to reduce the effects of global discrepancies due to scattering or imaging modality differences. This algorithm was evaluated by phantom studies while simulated tumors were controlled to move with various patterns in a complex humanoid torso. Approximately 15,000 phantom MV images were acquired at nine gantry angles, with different tumors moving within ranges between 10 and 20 mm. Tumors were successfully identified on every projection with a total maximum/average error of 1.84/0.98 mm. This algorithm was also applied to over 5,000 frames of MV projections acquired during radiation therapy of five lung cancer patients. This tumor-tracking methodology is capable of accurately locating lung tumors during treatment without implanting any internal fiducial markers nor delivering extra imaging radiation doses.

A key-point based real-time tracking of lung tumor in x-ray image sequence by using difference of Gaussians filtering and optical flow

In radiation therapy, for accurate radiation dose delivery to a target tumor and reduction of the extra exposure of normal tissues, real-time tumor tracking is typically an important technique in lung cancer treatment since lung tumors move with patients’ respiration. To observe a tumor motion in real time, x-ray fluoroscopic devices can be employed, and various tracking techniques have been proposed to track tumors. However, development of a fast and accurate tracking method for clinical use is still a challenging task since the obscured image of the tumor can cause decreased tracking accuracy and can result in additional processing time for remedying the accuracy. In this study, a new key-point-based tumor tracking method, which is sufficiently fast and accurate, is presented. Given an x-ray image sequence, the proposed method employs a difference-of-Gaussians filtering technique to detect key points in the tumor region of the first frame which are robust against noise and outliers in the subsequent frames. In the subsequent frames, these key points are tracked using a fast optical flow technique, and tumor motion is estimated via their movement. To evaluate the performance, the proposed method has been tested on several clinical kV and MV x-ray image sequences. The experimental results showed that the average of the root mean square errors of tracking were and for kV and MV x-ray image sequences, respectively. This tracking performance was more accurate than previous tracking methods. In addition, the average processing times for each frame were and for kV and MV image sequences, respectively, and the proposed method was faster than previous methods as well as shorter than frame acquisition interval. Therefore, the proposed method has the potential for both highly accurate and fast tumor tracking in clinical applications.

The Clinical and Dosimetric Impact of Real-Time Target Tracking in Pancreatic SBRT

Motion often hinders the safe delivery of ablative doses of radiation in the treatment of pancreatic tumors. Real-time tumor-tracking methods are an emerging technique to increase the accuracy of delivery. In this study, we report on a large, retrospective cohort of pancreatic patients treated with real-time, fiducial-based, kV-image guidance of stereotactic body radiation therapy (SBRT). The purpose of our study was to determine the impact of real-time tracking on treatment accuracy, tumor dose, and clinical workflow. Real-time tracking data from 68 patients treated with pancreatic SBRT were analyzed. The kV images orthogonal to the treatment beam were acquired in real time during treatment to visualize the location of implanted fiducial markers. Positional corrections were made if the fiducial markers were observed >3mm from the expected reference position. We recorded the frequency and nature of treatment interventions resulting from real-time tracking and derived a neural network-based dosimetric model to calculate the impact of these in-treatment interventions on target dose. Treatment pauses that required patient realignment because of real-time tumor tracking occurred during 32% of all fractions. The median magnitude of realignment shifts was 5.2mm (range, 2.1-18.9mm). Forty-five percent of shifts resulted in dosimetric differences to the tumor; of these, the median point dose difference was 23%±22% of prescription dose (maximum, 94%). The number of pauses per fraction was significantly higher in patients treated with respiratory gating (vs abdominal compression) and in patients with greater treatment time. Fiducial-based, real-time target tracking is clinically feasible for pancreatic SBRT treatment. Our data indicate that real-time tumor tracking leads to patient realignment in 32% of cases and results in significant benefits to target coverage. The increased accuracy of real-time target tracking may potentially enable safe dose escalation in pancreatic SBRT.

P266] Patient-optimized deep learning for robust tumor tracking

Purpose In markerless tumor tracking using X-ray fluoroscopic images for image-guided radiation therapy (IGRT), robustness against mistracking is important. Mistracking is primarily caused by bone structures projected on fluoroscopic images because they generally hide tumor features. We devised a real-time tumor-tracking method. The novelty of this method is to control “importance recognition” in computers, i.e., controlling recognition of image features that are important or unimportant for tumor tracking. This study demonstrates the feasibility of our method using clinical X-ray fluoroscopic images. Methods The strategy of our method was to utilize different positional co-occurrence probabilities in training images for supervised deep learning. A bone digitally reconstructed radiograph (DRR) was randomly overlapped on a soft-tissue DRR to create training images for deep learning, which established positional decorrelation between the bone feature and ground truth, indicating tumor position and shape. As the training of deep learning processed, this decorrelation induced the computer recognition that the bone feature was unimportant for tumor tracking. Moreover, this method was fully optimized for a patient because DRRs were created from patient-specific CT data only, unlike a general medical application of deep learning such as diagnosis that uses a massive amount of patient data. Results The accuracy of lung tumor tracking was within approximately 1 mm although the low-visible tumor and spine were overlapped on X-ray fluoroscopic images. Conclusions Feasibility of robust tumor tracking by our method was demonstrated. This method can be expected as a breakthrough technique that can contribute to the robust and safe implementation of markerless tumor tracking for radiation therapy in future.

Performance Evaluation of the Tumor Tracking Method Using Beam on/off Interface for the Treatment of Irregular Breathing

Dose rate regulated tracking is known to be an efficient method which adaptively delivers tracking treatments when patient breathing is irregular. The Motion Management Interface (MMI, Varian Medical System, CA), which provides beam on/off switching during treatment is available for clinic. Study is to test if delivering the adaptive tumor tracking is feasible for irregular breathing using beam switching with MMI. 55 free breathing RPM traces acquired from lung cancer patients are used. The first day RPM traces of the patients are utilized to design preprogrammed tracking MLC patterns, of which periods are intentionally reduced by 20% in order to catch up the variation of patient breathing irregularity in the treatment day. Eligibility criteria for this technique are the variation of amplitude and period less than 20%. An algorithm which determines beam on/off every 100 ms by considering the preprogrammed (MLC) positions and current breathing positions is developed. Tracking error and delivery efficacy are calculated by simulating the beam-switching adaptive tracking from the RPM traces. Breathing patterns of 38 patients (70%) met the eligibility criteria. Tracking errors of all of the cases who meet the criteria are less than 2 mm (average 1.4 mm) and the average delivery efficacy was 71%. Those of rest of the cases are 1.9 mm and 48%. Adaptive tracking with beam switching is feasible if patient selection is based on the eligibility criteria.

Markerless Respiratory Tumor Motion Prediction Using an Adaptive Neuro-fuzzy Approach

Accurate delivery of the prescribed dose to moving lung tumors is a key challenge in radiation therapy. Tumor tracking involves real-time specifying the target and correcting the geometry to compensate for the respiratory motion, that's why tracking the tumor requires caution. This study aims to develop a markerless lung tumor tracking method with a high accuracy. In this study, four-dimensional computed tomography (4D-CT) images of 10 patients were used, and all the slices which contained the tumor were contoured for all patients. The first four phases of 4D-CT images which contained tumors were selected as input of the software, and the next six phases were considered as the output. A hybrid intelligent method, adaptive neuro-fuzzy inference system (ANFIS), was used to evaluate motion of lung tumor. The root mean square error (RMSE) was used to investigate the accuracy of ANFIS performance for tumor motion prediction. For predicting the positions of contoured tumors, the averages of RMSE for each patient were calculated for all the patients. The results showed that the RMSE did not have a major variation. The data in the 4D-CT images were used for motion tracking instead of using markers that lead to more information of tumor motion with respect to methods based on marker location.

Feasibility of using single photon counting X-ray for lung tumor position estimation based on 4D-CT

PurposeIn stereotactic body radiation therapy of lung tumors, reliable position estimation of the tumor is necessary in order to minimize normal tissue complication rate. While kV X-ray imaging is frequently used, continuous application during radiotherapy sessions is often not possible due to concerns about the additional dose. Thus, ultra low-dose (ULD) kV X-ray imaging based on a single photon counting detector is suggested. This paper addresses the lower limit of photons to locate the tumor reliably with an accuracy in the range of state-of-the-art methods, i.e. a few millimeters. Method18 patient cases with four dimensional CT (4D-CT), which serves as a-priori information, are included in the study. ULD cone beam projections are simulated from the 4D-CTs including Poisson noise. The projections from the breathing phases which correspond to different tumor positions are compared to the ULD projection by means of Poisson log-likelihood (PML) and correlation coefficient (CC), and template matching under these metrics. ResultsThe results indicate that in full thorax imaging five photons per pixel suffice for a standard deviation in tumor positions of less than half a breathing phase. Around 50 photons per pixel are needed to achieve this accuracy with the field of view restricted to the tumor region. Compared to CC, PML tends to perform better for low photon counts and shifts in patient setup. Template matching only improves the position estimation in high photon counts. The quality of the reconstruction is independent of the projection angle. ConclusionsThe accuracy of the proposed ULD single photon counting system is in the range of a few millimeters and therefore comparable to state-of-the-art tumor tracking methods. At the same time, a reduction in photons per pixel by three to four orders of magnitude relative to commercial systems with flatpanel detectors can be achieved. This enables continuous kV image-based position estimation during all fractions since the additional dose to the patient is negligible.

高次局所自己相関特徴を用いた動的テンプレートマッチングによる超音波ガイド腫瘍追従手法

In this report we propose a novel robust tumor tracking method for ultrasound guided RFA treatments. RFA (radiofrequency ablation) treatments have a serious problem to hide a tumor to be treated due to the hyperechoic region, which is generated by vapors during RFA. Moreover, organ deformations seriously deteriorate the tracking performance. To cope with these problems, we propose a novel method to track a tumor in ultrasound diagnostic image. Templates, which is used to track the tumor, are generated dynamically by HLAC(Higher order Local Auto Correlation) -based subspace method. Our method achieves stable tracking by selecting templates automatically based on the texture features of ultrasound diagnostic images, while the conventional method is unstable due to the variation to select templates manually. Experimental results show the effectiveness of our proposing motion tracking method concerning the robustness and accuracy.

Markerless tumor tracking using short kilovoltage imaging arcs for lung image-guided radiotherapy

The ability to monitor tumor motion without implanted markers is clinically advantageous for lung image-guided radiotherapy (IGRT). Existing markerless tracking methods often suffer from overlapping structures and low visibility of tumors on kV projection images. We introduce the short arc tumor tracking (SATT) method to overcome these issues. The proposed method utilizes multiple kV projection images selected from a nine-degree imaging arc to improve tumor localization, and respiratory-correlated 4D cone-beam CT (CBCT) prior knowledge to minimize the effects of overlapping anatomies. The 3D tumor position is solved as an optimization problem with prior knowledge incorporated via regularization. We retrospectively validated SATT on 11 clinical scans from four patients with central tumors. These patients represent challenging scenarios for markerless tumor tracking due to the inferior adjacent contrast. The 3D trajectories of implanted fiducial markers were used as the ground truth for tracking accuracy evaluation. In all cases, the tumors were successfully tracked at all gantry angles. Compared to standard pre-treatment CBCT guidance alone, trajectory errors were significantly smaller with tracking in all cases, and the improvements were the most prominent in the superior-inferior direction. The mean 3D tracking error ranged from 2.2–9.9 mm, which was 0.4–2.6 mm smaller compared to pre-treatment CBCT. In conclusion, we were able to directly track tumors with inferior visibility on kV projection images using SATT. Tumor localization accuracies are significantly better with tracking compared to the current standard of care of lung IGRT. Future work involves the prospective evaluation and clinical implementation of SATT.

WE-AB-303-03: A Scheme for Real-Time, Marker-Less and Low-Dose Tumor Tracking Using Scattering Imaging: Monte Carlo Simulation Study

Purpose: In-treatment tumor localization is critical for the management of tumor motion in lung cancer radiotherapy. Conventional tumor-tracking methods using a kV or MV x-ray projection has limited contrast. To facilitate real-time, marker-less and low-dose in-treatment image tumor tracking, we propose a novel scheme using Compton scatter imaging. This study reports Monte Carlo (MC) simulations on this scheme for the purpose of proof-of-principle. Methods: A slit x-ray beam along the patient superior-inferior (SI) direction is directed to the patient, intersecting the patient lung at a 2D plane containing majority part of the tumor motion trajectory. X-ray photons are scattered due to Compton effect from this plane, which are spatially collimated by, e.g., a pinhole, on one side of the plane and then captured by a detector behind it. The captured image, after correcting for x-ray attenuation and scatter angle variation, reflects the electron density, which allows visualization of the instantaneous anatomy on this plane. We performed MC studies on a phantom and a patient case for the initial test of this proposed method. Results: In the phantom case, the contrast-resolution calculated using tumor/lung as foreground/background for kV fluoroscopy, cone-beam CT, and scattering image were 0.0625, 0.6993, and 0.5290, respectively. In the patient case, tumor motion can be clearly observed in the scatter images. Compared to fluoroscopy, scattering imaging also significantly reduced imaging dose because of its narrower beam design. Conclusion: MC simulation studies demonstrated the potential of the proposed scheme in terms of capturing the instantaneous anatomy of a patient on a 2D plane. Clear visualization of the tumor will probably facilitate ‘marker-less’ and ‘real-time’ tumor tracking with low imaging dose. NIH (1R01CA154747-01, 1R21CA178787-01A1 and 1R21EB017978-01A1)

Surrogate-driven deformable motion model for organ motion tracking in particle radiation therapy

The aim of this study is the development and experimental testing of a tumor tracking method for particle radiation therapy, providing the daily respiratory dynamics of the patient’s thoraco-abdominal anatomy as a function of an external surface surrogate combined with an a priori motion model. The proposed tracking approach is based on a patient-specific breathing motion model, estimated from the four-dimensional (4D) planning computed tomography (CT) through deformable image registration. The model is adapted to the interfraction baseline variations in the patient’s anatomical configuration. The driving amplitude and phase parameters are obtained intrafractionally from a respiratory surrogate signal derived from the external surface displacement. The developed technique was assessed on a dataset of seven lung cancer patients, who underwent two repeated 4D CT scans. The first 4D CT was used to build the respiratory motion model, which was tested on the second scan. The geometric accuracy in localizing lung lesions, mediated over all breathing phases, ranged between 0.6 and 1.7 mm across all patients. Errors in tracking the surrounding organs at risk, such as lungs, trachea and esophagus, were lower than 1.3 mm on average. The median absolute variation in water equivalent path length (WEL) within the target volume did not exceed 1.9 mm-WEL for simulated particle beams. A significant improvement was achieved compared with error compensation based on standard rigid alignment. The present work can be regarded as a feasibility study for the potential extension of tumor tracking techniques in particle treatments. Differently from current tracking methods applied in conventional radiotherapy, the proposed approach allows for the dynamic localization of all anatomical structures scanned in the planning CT, thus providing complete information on density and WEL variations required for particle beam range adaptation.

Investigation of the robustness of adaptive neuro-fuzzy inference system for tracking moving tumors in external radiotherapy.

In external radiotherapy of dynamic targets such as lung and breast cancers, accurate correlation models are utilized to extract real time tumor position by means of external surrogates in correlation with the internal motion of tumors. In this study, a correlation method based on the neuro-fuzzy model is proposed to correlate the input external motion data with internal tumor motion estimation in real-time mode, due to its robustness in motion tracking. An initial test of the performance of this model was reported in our previous studies. In this work by implementing some modifications it is resulted that ANFIS is still robust to track tumor motion more reliably by reducing the motion estimation error remarkably. After configuring new version of our ANFIS model, its performance was retrospectively tested over ten patients treated with Synchrony Cyberknife system. In order to assess the performance of our model, the predicted tumor motion as model output was compared with respect to the state of the art model. Final analyzed results show that our adaptive neuro-fuzzy model can reduce tumor tracking errors more significantly, as compared with ground truth database and even tumor tracking methods presented in our previous works.

SU‐E‐J‐156: Preclinical Inverstigation of Dynamic Tumor Tracking Using Vero SBRT Linear Accelerator: Motion Phantom Dosimetry Study

Purpose:Following the ‘end‐to‐end testing’ paradigm of Dynamic Target Tracking option in our Image‐Guided dedicated SBRT VeroTM linac, we verify the capability of the system to deliver planned dose to moving targets in the heterogeneous thorax phantom (CIRSTM). The system includes gimbaled C‐band linac head, robotic 6 degree of freedom couch and a tumor tracking method based on predictive modeling of target position using fluoroscopically tracked implanted markers and optically tracked infrared reflecting external markers.Methods:4DCT scan of the motion phantom with the VisicoilTM implanted marker in the close vicinity of the target was acquired, the ‘exhale’=most prevalent phase was used for planning (iPlan by BrainLabTM). Typical 3D conformal SBRT treatment plans aimed to deliver 6‐8Gy/fx to two types of targets: a)solid water‐equivalent target 3cm in diameter; b)single VisicoilTM marker inserted within lung equivalent material. The planning GTV/CTV‐to‐PTV margins were 2mm, the block margins were 3 mm. The dose calculated by MonteCarlo algorithm with 1% variance using option Dose‐to‐water was compared to the ion chamber (CC01 by IBA Dosimetry) measurements in case (a) and GafchromicTM EBT3 film measurements in case (b). During delivery, the target 6 motion patterns available as a standard on CIRSTM motion phantom were investigated: in case (a), the target was moving along the designated sine or cosine4 3D trajectory; in case (b), the inserted marker was moving sinusoidally in 1D.Results:The ion chamber measurements have shown the agreement with the planned dose within 1% under all the studied motion conditions. The film measurements show 98.1% agreement with the planar calculated dose (gamma criteria: 3%/3mm).Conclusion:We successfully verified the capability of the SBRT VeroTM linac to perform real‐time tumor tracking and accurate dose delivery to the target, based on predictive modeling of the correlation between implanted marker motion and external surrogate of breathing motion.

Accurate tumor localization and tracking in radiation therapy using wireless body sensor networks

Radiation therapy is an effective method to combat cancerous tumors by killing the malignant cells or controlling their growth. Knowing the exact position of the tumor is a very critical prerequisite in radiation therapy. Since the position of the tumor changes during the process of radiation therapy due to the patient׳s movements and respiration, a real-time tumor tracking method is highly desirable in order to deliver a sufficient dose of radiation to the tumor region without damaging the surrounding healthy tissues.In this paper, we develop a novel tumor positioning method based on spatial sparsity. We estimate the position by processing the received signals from only one implantable RF transmitter. The proposed method uses less number of sensors compared to common magnetic transponder based approaches. The performance of the proposed method is evaluated in two different cases: (1) when the tissue configuration is perfectly determined (acquired beforehand by MRI or CT) and (2) when there are some uncertainties about the tissue boundaries. The results demonstrate the high accuracy and performance of the proposed method, even when the tissue boundaries are imperfectly known.

69: Intra-fraction tumor tracking based on a surrogate-driven 4D CT motion model in particle radiation therapy

Purpose: The application of tumor tracking techniques for intra-fraction motion compensation in particle radiation therapy is challenging due to the need to account both for target motion and for particle range fluctuations induced by tissue density variations along the beam line. The aim of this work is to develop and investigate a tumor tracking method for an application in particle therapy, providing the daily respiratory dynamics of the entire patient anatomy as a function of an external surface surrogate combined with an a priori motion model.

Tumor Tracking Method Based on a Deformable 4D CT Breathing Motion Model Driven by an External Surface Surrogate

To develop a tumor tracking method based on a surrogate-driven motion model, which provides noninvasive dynamic localization of extracranial targets for the compensation of respiration-induced intrafraction motion in high-precision radiation therapy. The proposed approach is based on a patient-specific breathing motion model, derived a priori from 4-dimensional planning computed tomography (CT) images. Model parameters (respiratory baseline, amplitude, and phase) are retrieved and updated at each treatment fraction according to in-room radiography acquisition and optical surface imaging. The baseline parameter is adapted to the interfraction variations obtained from the daily cone beam (CB) CT scan. The respiratory amplitude and phase are extracted from an external breathing surrogate, estimated from the displacement of the patient thoracoabdominal surface, acquired with a noninvasive surface imaging device. The developed method was tested on a database of 7 lung cancer patients, including the synchronized information on internal and external respiratory motion during a CBCT scan. About 30 seconds of simultaneous acquisition of CBCT and optical surface images were analyzed for each patient. The tumor trajectories identified in CBCT projections were used as reference and compared with the target trajectories estimated from surface displacement with the a priori motion model. The resulting absolute differences between the reference and estimated tumor motion along the 2 image dimensions ranged between 0.7 and 2.4 mm; the measured phase shifts did not exceed 7% of the breathing cycle length. We investigated a tumor tracking method that integrates breathing motion information provided by the 4-dimensional planning CT with surface imaging at the time of treatment, representing an alternative approach to point-based external-internal correlation models. Although an in-room radiograph-based assessment of the reliability of the motion model is envisaged, the developed technique does not involve the estimation and continuous update of correlation parameters, thus requiring a less intense use of invasive imaging.