Click here to

Session: Therapy: Treatment Verification [Return to Session]

Evaluation of Dynamic Tumour Tracking (DTT) uncertainty for a Fiducial-Based Stereotactic Ablative Radiotherapy (SABR) Treatment of the Liver Using EPID Images

M Rostamzadeh*, A Bergman, K Luchka, BC Cancer, Vancouver, BC, CA


TH-IePD-TRACK 5-5 (Thursday, 7/29/2021) 3:00 PM - 3:30 PM [Eastern Time (GMT-4)]

Purpose: To assess dynamic tumour tracking (DTT) uncertainty for fiducial-based stereotactic ablative radiotherapy (SABR) treatments of the liver using electronic portal imaging device (EPID) images.

Methods: EPID images were acquired during 3D conformal radiotherapy (3DCRT) DTT using a Vero4DRT linac (Brainlab-Mitsubishi), equipped with an amorphous-silicon (aSi) EPID (maximum frame-rate: 2Hz). Images acquired in phantom and patients were pre-processed using a Laplacian-of-Gaussian (LoG) function. Three gold fiducials were inserted into an acrylic phantom. The phantom was placed on a 1D, moving Brainlab platform with sine-wave motion synchronized to a respiratory platform. EPID images were acquired for 3DCRT noncoplanar DTT beams. The EPID images of nine liver SABR patients treated with 3DCRT noncoplanar beams were also analyzed. All patients had 3-4 implanted fiducials. A chain-code algorithm was applied to the EPID images to detect field-edges of the MLC-defined apertures. Fiducials were detected using a connected-neighbor algorithm. For each beam, the centre-of-mass (COM) of the fiducials relative to the field-edge was reported (at isocentre plane).

Results: Of the 123 EPID phantom images, 100% of the fiducials were located. The Standard-Deviation (SD) of the relative position between the fiducial COM and field-edge in the X(h) and Y(v) direction, for all fields, was 0.2 and 0.3 mm, respectively. Of the 1418 EPID patient images acquired, 81% had detectable fiducials. Fiducials may not be detected if the images were noisy, or the location was occluded by the MLC. The SD of the relative position between the fiducial COM and field-edge in X and Y direction was between 0.27-0.82 mm and 0.53-1.61mm, respectively. Sources of uncertainty in the relative positions include field-edge localization, fiducial-localization, and physical tracking uncertainty during treatment.

Conclusion: DTT uncertainty can be evaluated on a field-by-field basis using EPID images. The DTT uncertainty was <0.82 and 1.61mm in X and Y direction, respectively.



    Not Applicable / None Entered.


    Not Applicable / None Entered.

    Contact Email