ESTRO 2025 - Abstract Book

S3059

Physics - Image acquisition and processing

ESTRO 2025

BD MG and BD CT presented an acceptable agreement (within 20%) in 85% of patients and a good agreement (within 15%) in 65% (Fig 1). Two outliers were identified, where visual inspection suggested potential over-estimation by LIBRA on the MG (Fig. 2). Excluding these outliers, BD CT was consistently higher than BD MG (on average by 11.4%). Classification into BIRADS categories resulted in 60% exact agreement between CT and MG (i.e. same category) and 90% one-category difference, aligning with reported inter-observer variability[5]. Conclusion: Comparing CT and MG-calculated density is non-trivial due to inherent modality differences, but results of this feasibility study are encouraging, suggesting that BD estimation from CT scans is comparable to established methods on MGs. Poor agreement could reflect our younger population and will be further investigated in a cohort of 200 patients. References: [1] Boyd et al: Mammographic density and the risk and detection of breast cancer. N Engl J Med 2007. [2] Keller et al. Estimation of breast percent density in raw and processed full field digital mammography images [...]. Medical physics. 2012. [3] Gastounioti et al: Evaluation of LIBRA software for fully automated mammographic density assessment in breast cancer risk prediction. Radiology 2020. [4] Melnikow et al: A Systematic Review for the U.S. Preventive Service Task Force. Table 1. https://www.ncbi.nlm.nih.gov/books/NBK343794/table/ch1.t1/ [5] Redondo et al: Inter- and intraradiologist variability in the BI-RADS assessment and breast density categories for screening mammograms, British Journal of Radiology , 2012. Keywords: Breast Density, Breast Radiotherapy, Novel Method Digital Poster Qualitative assessment of HyperSight CBCT image quality using breath hold technique for image matching and contouring Hannah Marcelle Dahn 1 , Amanda Cherpak 1,2 , Robert Lee Macdonald 1,2 , Natasha McMaster 1 , Kenny Zhan 1 , Slawa Cwajna 1 , Nikhilesh Patil 1 , James Robar 1,2 1 Radiation Oncology, QE2 Cancer centre, Halifax, Canada. 2 Department of Physics and Atmospheric Science, Dalhousie University, Halifax, Canada Purpose/Objective: The intent of this study was to qualitatively assess the image quality provided by a cone beam computed tomography (CBCT) platform (HyperSight, Varian Medical Systems). Contouring confidence was assessed compared to fan beam CT (FBCT) acquired on a CT simulator and confidence in image matching was compared to conventional CBCT. Material/Methods: This study included 30 patients treated under breath hold for tumors in the lung, breast, and upper abdomen (liver, pancreas, and peri-aortic lymph nodes). The patients were imaged with CBCT on a standard image guidance platform (TB CBCT) (TrueBeam 2.7, Varian Medical Systems), FBCT (CTsim, GE Optima) and HyperSight imaging (HS CBCT). Confidence in image matching was graded by Radiation Therapists (RTTs) (n=5) for TB CBCT vs HS CBCT on a scale from 1-5 (1 = much more confident with TB CBCT, 3 = same confidence and 5 = much more confident with HS CBCT). Confidence in contouring specific structures (esophagus (lung cases), adjacent bowel (upper abdomen cases) and heart and target structures (breast cases)) was graded by Radiation Oncologists (ROs) (n=4) on a scale from 1-5 (1= very low to 5 = very high) for HS CBCT and FBCT. 3621