Virtual Library
Start Your Search
P. Keall
Author of
-
+
MA13 - Modern Technologies and Biological Factors in Radiotherapy (ID 395)
- Event: WCLC 2016
- Type: Mini Oral Session
- Track: Radiotherapy
- Presentations: 1
- Moderators:M. Thomas, P. Mitchell
- Coordinates: 12/06/2016, 16:00 - 17:30, Stolz 1
-
+
MA13.01 - Markerless Tumour Tracking during Lung Radiotherapy Using Intrafraction X-Ray Imaging (ID 5533)
16:00 - 16:06 | Author(s): P. Keall
- Abstract
- Presentation
Background:
Lung tumours often exhibit large and unpredictable motion that can severely compromise radiotherapy outcomes. Markerless tumour tracking can enable wide access to motion-adaptive radiotherapy, negating the risks and costs associated with implanting markers. The main barrier to markerless tumour tracking is the inferior tumor visibility on x-ray images due to overlapping anatomic structures. The aim of this study is to develop a markerless tumor tracking method for lung radiotherapy using intrafraction x-ray imaging.
Methods:
The markerless tumour tracking method (Figure1a) consists of four steps: (1) Building a tumour and anatomic model from the cone-beam CT (CBCT) acquired prior to treatment, (2) Using the anatomic model to remove the contribution of anatomic structures on intrafraction x-ray images, (3) Locating the tumour on the intrafraction 2D x-ray image via template matching using the tumour model, (4) Determining the tumour 3D position by a Kalman filter. The proposed method was retrospectively validated on (i) 11 CBCT scans from four patients with central tumours, and (ii) a kV fluoroscopic scan during a stereotactic ablative radiotherapy (SABR) treatment from the Light SABR trial (NCT02514512). Tracking errors were estimated using the motions of markers or beacons implanted near the tumours. Figure 1
Results:
Markerless tumour tracking successfully tracked tumours in all cases at every imaging angle. The mean 3D tracking error ranged from 1.8-4.1mm for the 11 CBCT scans, and was 3.0mm for the SABR case. Compared with the current standard of care, i.e. a single estimation of tumour position prior to treatment from the pre-treatment CBCT, markerless tumour tracking reduced tumour localization error by 0.9-7.9mm. Tracking errors in the left-right, superior-inferior, and anterior-posterior directions are shown in Figure1b.
Conclusion:
A markerless tumour tracking method was developed and shown to improve tumour localization accuracy in 12 lung cancer cases. This method can potentially enable wide access to motion-adaptive radiotherapy.
Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.
-
+
P2.05 - Poster Session with Presenters Present (ID 463)
- Event: WCLC 2016
- Type: Poster Presenters Present
- Track: Radiotherapy
- Presentations: 1
- Moderators:
- Coordinates: 12/06/2016, 14:30 - 15:45, Hall B (Poster Area)
-
+
P2.05-039 - Assessment of Lung Tumour Motion Comparing 4DCT, 4DCBCT and Motion of Implanted Beacons during Imaging and Irradiation (ID 5101)
14:30 - 14:30 | Author(s): P. Keall
- Abstract
Background:
Moving lung tumours exceeding the observed motion from planning 4D computed tomography (4DCT) can result in reduced dose coverage in stereotactic ablative body radiation therapy (SABR). 4D cone-beam CT (4DCBCT) facilitates verification of tumour trajectories before each treatment fraction. Using implanted Calypso beacons in the lung as ground truth, this work aims to assess how well 4DCT and 4DCBCT represent the actual motion range during imaging and irradiation.
Methods:
4DCBCT was reconstructed for 1-2 fractions of 6 patients (three implanted Calypso beacons) receiving lung SABR from the projections acquired for treatment setup CBCT. Two reconstructions per projection set were created using the prior image constrained compressed sensing (PICCS) method based on the Calypso motion trajectories or an external respiratory signal (Philips Bellows). Calypso beacons were segmented for all 10 bins of the 4DCT and 4DCBCT sets and the centroid position calculated. Beacon centroid motion as seen on the 4DCT and 4DCBCT with respect to reference phase (end-exhale) was extracted and compared with the actual beacon centroid motion during CBCT acquisition and during irradiation.
Results:
Both methods for 4DCBCT reconstruction failed to capture sudden motion peaks during scanning (see Fig. 1), but performed similar to the 4DCT. In general, 4DCT and 4DCBCT underestimated the actual beacon centroid motion. In the SI direction 22-27% of the actual motion exceeded the motion range from 4DCT and 4DCBCT imaging. In AP and LR direction up to 39-58% of the motion exceeded the observed motion range from 4D imaging. Figure 1
Conclusion:
Both 4DCT and 4DCBCT failed to represent the full tumour motion range. For a safe treatment delivery this needs to be accounted for either by sufficient margins or more preferably real-time treatment adaptation directly tackling motion peaks and unpredictable motion.
