Click here to

Session: Imaging: CT Artifacts, Reconstruction, and Imaging Performance Assessment [Return to Session]

CARE-CT: CArdiac and REspiratory Adaptive Computed Tomography

N Morton*, T Reynolds, P Keall, R O'Brien, University of Sydney, Eveleigh, NSWAU,


TU-IePD-TRACK 1-2 (Tuesday, 7/27/2021) 3:00 PM - 3:30 PM [Eastern Time (GMT-4)]

Purpose: To safely plan high dose radiotherapy treatments for thoracic cancers and novel treatments such as non-invasive cardiac radioablation, 4DCT requires the capacity to account for multiple motion sources. We have developed a novel acquisition method, CARE-CT (CArdiac and REspiratory adaptive Computed Tomography), to gate for dual motion and quantified the reduction in image artifacts.

Methods: CARE-CT and conventional cine 4DCT imaging were simulated on a deformable digital phantom (XCAT) for nineteen patient-measured ECG and respiratory traces. Our software monitored motion in real-time, triggering a simulated CT acquisition during combined cardiac and respiratory phases with 0.36s gantry rotation and 2cm/1s couch translation. Six and ten respiratory phases during cardiac diastole (60-80% through cardiac cycle) were acquired for each scan. Image artifacts were quantified and compared for 13 thoracic substructures using the normalized cross correlation between axial slices (ΔNCC). The 13 substructure volumes were compared to an average XCAT ground truth and compared between protocols. Total scan and beam-on times were recorded.

Results: Compared to conventional 4DCT, the pooled mean reduction in the ΔNCC was 19% and 20% for the 6 and 10 bin CARE-CT scans respectively, with the largest reduction of 37% for the left ventricular wall. The mean variation in total structure volumes compared to the ground truth was reduced from 3% (-19% to 59%) for conventional 4DCT to 0% (-1% to 2%) for CARE-CT. The mean total scan times were increased from 1.8±0.4min for conventional 4DCT to 3.5±0.9min and 7.5±2.2min for 6 and 10 bin scans, respectively. The total beam on time was reduced from 98±28s for conventional 4DCT to 27±0.2s and 45±0.6s for 6 and 10 bin scans, respectively.

Conclusion: CARE-CT reduced image artifacts in thoracic substructures while reducing beam-on time as compared. This study provides a crucial first step towards dual-motion management for central tumours.

Funding Support, Disclosures, and Conflict of Interest: R. O'Brien would like to acknowledge that this project was supported by Cancer Australia PdCCRS grants numbers 1161748 and 1123068.



    Not Applicable / None Entered.


    Not Applicable / None Entered.

    Contact Email