Click here to


Are you sure ?

Yes, do it No, cancel

First Clinical Implementation of Respiratory Adaptive 4D-CT (REACT)

N Morton1*, J Barber2 , J Sykes3 , J Kipritidis4 , C Hofmann5 , P Keall6 , R O'Brien7 , (1) University of Sydney, Sydney, NSW, (2) BLACKTOWN CANCER AND HAEMATOLOGY CENTRE, Sydney, NSW, (3) Blacktown Hospital, Blacktown, ,(4) Northern Sydney Cancer Centre, Sydney, ,(5) Siemens Healthcare GmbH, Forchheim, Bavaria, (6) University of Sydney, Camperdown, NSW, (7) University of Sydney, Camperdown, NSW


(Sunday, 7/29/2018) 3:00 PM - 3:30 PM

Room: Exhibit Hall | Forum 8

Purpose: Respiratory Adaptive Computed Tomography (REACT) system aims to mitigate motion errors which currently occur in up to 90% of 4DCT images. Errors are caused by inconsistent patient breathing, and have deleterious effects on tumor delineation and functional imaging. The REACT system identifies changes to a patient’s breathing cycle and gates the beam in real-time. A method to implement REACT on a clinical CT scanner for the first time was developed and its effect on artifact reduction quantified.

Methods: (1) Three free-breathing respiratory traces of high, average and low irregularity were run in-silico through our in-house REACT software. This resulted in a trace with ‘Beam On’/‘Beam Off’ flags. 2) ‘Beam Off’ segments of each respiratory trace were retrospectively removed, leaving only regions of regular breathing. 3) Manipulated traces were simulated on CIRS dynamic thorax phantom and images acquired using standard 4DCT helical protocol (pitch=0.09). 4) Each set of images were compared to conventional 4DCT images through visual inspection, total volume, false positive(FP) and false negative(FN) rates of the segmented moving phantom insert.

Results: For the first time a phantom method simulating the integration of REACT system with a clinical scanner was successfully developed and real CT images acquired. All three traces ran without error. Image artifacts were found in all conventional scans. Total volume error was 28% for highly irregular trace which showed high FN rate of 25%, supporting severe overlapping visually inspected in images. Quantification of image artifact reduction is currently underway.

Conclusion: First-time experimental implementation of REACT system and a clinical CT scanner was successful. Severe artifacts were found in conventional scans and continuing analysis of REACT images should demonstrate artifact reduction. This study represents the first step to complete implementation of a REACT real-time triggered CT system that should reduce not only imaging errors but imaging dose.

Funding Support, Disclosures, and Conflict of Interest: This work was supported in part by funding from Siemens Healthcare.


Not Applicable / None Entered.



Contact Email