Purpose: Prior to thoracic radiotherapy treatments, patients are commonly aligned using conventional 4DCBCT scans (1320 projections over 4 minutes). We determine the viability of rapid low dose 4DCBCT scans by simulating scans as low as 60 projections over 25 seconds, using data from 20 patients of the ADAPT clinical trial (NCT04070586).
Methods: The ADAPT clinical trial performs adaptive 4DCBCT scans by adapting gantry rotational speed and projection acquisition in accordance to real-time patient breathing rate changes, this produces 600 and 200 projection scans in a quarter of conventional scan time. In this study, these adaptive 4DCBCT scans were down sampled and simulated into even faster and lower imaging dose 310, 160, 110 and 60 projection. Reconstructed image quality was quantified using Structural Similarity Index (SSIM), Signal-to-Noise-Ratio (SNR), Contrast-to-Noise-Ratio (CNR), Tissue Interface Sharpness-Diaphragm (TIS-D) and Tissue Interface Sharpness-Tumor (TIS-T). Conventional 4DCBCT scans were used as a comparator.
Results: The simulated 310, 160, 110 and 60 projection adaptive acquisitions deliver approximately (76%, 87%, 91%, 95%) reduction in imaging dose via reduction in total number of projections acquired and (48%, 73%, 82%, 90%) shorter scan time on average compared to conventional 4DCBCT scans, respectively. The 310, 160 and 110 projection adaptive acquisitions from all patients matched or improved compared to the conventional acquisition across all image quality metrics. The minimum number of projections to maintain the image quality metrics of conventional scans (SSIM=1±0, SNR=11.9±11.7, CNR=9.5±3.1, TIS-D=0.59±0.3 and TIS-T=0.41±0.1) was the 110 projection adaptive scan (SSIM=0.98±0.01, SNR=18.5±8.9, CNR=10.2±3.5, TIS-D=0.72±0.4, TIS-T=0.57±0.7).
Conclusion: 110 projection adaptive acquisition results show image quality can be maintained comparable to conventional acquisition whilst reducing scan time by 95% and dose by 90%.
Funding Support, Disclosures, and Conflict of Interest: This study is funded by NHMRC grant 1138899 and partly by a Cancer Australia grant number 1123068. RO is supported by a Cancer Institute NSW career development fellowship, and PK is supported by an Australian NHMRC Senior Principal Research Fellowship.