Fast image acquisition and processing on a TV camera-based portal imaging system

Suitability of markerless EPID tracking for tumor position verification in gated radiotherapy

PURPOSE

To maximize the benefits of respiratory gated radiotherapy (RGRT) of lung tumors real-time verification of the tumor position is required. This work investigates the feasibility of markerless tracking of lung tumors during beam-on time in electronic portal imaging device (EPID) images of the MV therapeutic beam.

METHODS

EPID movies were acquired at ∼2 fps for seven lung cancer patients with tumor peak-to-peak motion ranges between 7.8 and 17.9 mm (mean: 13.7 mm) undergoing stereotactic body radiotherapy. The external breathing motion of the abdomen was synchronously measured. Both datasets were retrospectively analyzed in PortalTrack, an in-house developed tracking software. The authors define a three-step procedure to run the simulations: (1) gating window definition, (2) gated-beam delivery simulation, and (3) tumor tracking. First, an amplitude threshold level was set on the external signal, defining the onset of beam-on/-off signals. This information was then mapped onto a sequence of EPID images to generate stamps of beam-on/-hold periods throughout the EPID movies in PortalTrack, by obscuring the frames corresponding to beam-off times. Last, tumor motion in the superior-inferior direction was determined on portal images by the tracking algorithm during beam-on time. The residual motion inside the gating window as well as target coverage (TC) and the marginal target displacement (MTD) were used as measures to quantify tumor position variability.

RESULTS

Tumor position monitoring and estimation from beam's-eye-view images during RGRT was possible in 67% of the analyzed beams. For a reference gating window of 5 mm, deviations ranging from 2% to 86% (35% on average) were recorded between the reference and measured residual motion. TC (range: 62%-93%; mean: 77%) losses were correlated with false positives incidence rates resulting mostly from intra-/inter-beam baseline drifts, as well as sudden cycle-to-cycle fluctuations in exhale positions. Both phenomena can lead to considerable deviations (with MTD values up to a maximum of 7.8 mm) from the intended tumor position, and in turn may result in a marginal miss. The difference between tumor traces determined within the gating window against ground truth trajectory maps was 1.1 ± 0.7 mm on average (range: 0.4-2.3 mm).

CONCLUSIONS

In this retrospective analysis of motion data, it is demonstrated that the system is capable of determining tumor positions in the plane perpendicular to the beam direction without the aid of fiducial markers, and may hence be suitable as an online verification tool in RGRT. It may be possible to use the tracking information to enable on-the-fly corrections to intra-/inter-beam variations by adapting the gating window by means of a robotic couch.

Three-dimensional spatial modelling of the correlation between abdominal motion and lung tumour motion with breathing

The aim of this research was to investigate whether a spatial correlation could be found between an external 3-D respiratory signal and the tumour trajectory. The respiratory signal was obtained by tracking the abdominal movement and the tumour trajectory was obtained by automatically determining the tumour position in a series of portal images. Three different models, based on Systems Identification, are presented to model the correlation using a 1-D respiratory signal, a 3-D respiratory signal and a 3-D respiratory signal together with previously determined tumour positions. Adequate correlation was found for all models in the direction of the tumour movement with standard deviations of 0.89 mm, 0.72 mm and 0.75 mm, respectively, and model fit of Rt2 = 0.19, 0.63 and 0.82, respectively. Increasing the frame rate for the acquisition of portal images from 3 to 15 frames per second improved the standard deviation and model fit. In summary, it is possible to spatially correlate a 3-D respiratory signal with the tumour trajectory using this approach. The models presented provide a framework that can be extended to include more information if required. A 3-D respiratory signal is preferable to a 1-D signal in modelling the tumour motion that is not along the main axis of tumour movement.

Clinical Feasibility of Using an EPID in cine Mode for Image-Guided Verification of Stereotactic Body Radiotherapy

PURPOSE

To introduce a novel method for monitoring tumor location during stereotactic body radiotherapy (SBRT) while the treatment beam is on by using a conventional electronic portal imaging device (EPID).

METHODS AND MATERIALS

In our clinic, selected patients were treated under a phase I institutional review board-approved SBRT protocol for limited hepatic metastases from solid tumors. Before treatment planning multiple gold fiducial markers were implanted on the periphery of the tumor. During treatment the EPID was used in cine mode to collect the exit radiation and produce a sequence of images for each field. An in-house program was developed for calculating the location of the fiducials and their relative distance to the planned locations.

RESULTS

Three case studies illustrate the utility of the technique. Patient A exhibited a systematic shift of 4 mm during one of the treatment beams. Patient B showed an inferior drift of the target of approximately 1 cm from the time of setup to the end of the fraction. Patient C had a poor setup on the first day of treatment that was quantified and accounted for on subsequent treatment days.

CONCLUSIONS

Target localization throughout each treatment beam can be quickly assessed with the presented technique. Treatment monitoring with an EPID in cine mode is shown to be a clinically feasible and useful tool.

Tracking moving objects with megavoltage portal imaging: A feasibility studya)

Four different algorithms were investigated with the aim to determine their suitability to track an object in conventional megavoltage portal images. The algorithms considered were the mean of the sum of squared differences (MSSD), mutual information (MI), the correlation ratio (CR), and the correlation coefficient (CC). Simulation studies were carried out with various image series containing a rigid object of interest that was moved along a predefined trajectory. For each of the series the signal-to-noise ratio (SNR) was varied to compare the performance of the algorithms under noisy conditions. For a poor SNR of -6 dB the mean tracking error was 2.4, 6.5, 39.0, and 17.2 pixels for MSSD, CC, CR and MI, respectively, with a standard deviation of 1.9, 12.9, 19.5, and 7.5 pixels, respectively. The size of a pixel was 0.5 mm. These results improved to 1.1, 1.3, 1.3, and 2.0 pixels, respectively, with a standard deviation of 0.6, 0.8, 0.8, and 2.1 pixels, respectively, when a mean filter was applied to the images prior to tracking. The implementation of MSSD into existing in-house software demonstrated that, depending on the search range, it was possible to process between 2 and 15 images/s, making this approach capable of real-time applications. In conclusion, the best geometric tracking accuracy overall was obtained with MSSD, followed by CC, CR, and MI. The simplest and best algorithm, both in terms of geometric accuracy as well as computational cost, was the MSSD algorithm and was therefore the method of choice.