Exhibit Hall | Forum 8
Purpose: To investigate the feasibility of using Tin filter in routine abdominal CT for regular and large size patients.
Methods: An abdominal phantom (30cmx40cm) with additional gel pads was used to represent abdomen of regular, large, very large and bariatric patients, with effective diameters of 35.9cm, 39.2cm, 44.5cm, and 52.2cm, respectively. Each phantom setting was scanned at Siemens Force CT scanner with 4 protocols: clinical abdomen protocol (CarekV on, 120 Reference kV, QRM 147mAs), 120kV (matched clinical protocol CTDI), 100Sn (100kV with 0.6mm Tin filter, matched clinical protocol CTDI or maximum output), 150Sn (150kV with 0.6mm Tin filter, matched clinical protocol CTDI or maximum output). For very large and bariatric sizes, certain protocol was not used due to substantial artifacts or output limits. All images were reconstructed with ADMIRE-3, Br40, and 3mm thickness. Noise was measured from 6 different locations of water background and contrast signal was measured with 15mg/mL Iodine inserts, and. To compare protocols for each phantom size, CTDI was normalized to identical noise level (for non-contrast scan), and identical CNR level (for contrast scan), with baseline of 150Sn protocol.
Results: 100Sn protocol is able to achieve clinical dose level for regular size patients, while is limited for large to bariatric patients due to tube power. At the same noise level, 100Sn showed >75% more dose efficiency than non-Tin clinical protocol for regular and large size patients; 150Sn showed >58% more dose efficiency than non-Tin clinical protocol. At the same CNR level, non-Tin protocols outperform both Tin protocols for regular to very large patients, while for bariatric patients, 150Sn has the highest CNR values compared to clinical 150kV.
Conclusion: Tin filter can substantially improve dose efficiency for non-contrast abdominal CT. For contrast abdominal CT of bariatric patients, 150Sn outperforms all non-Tin protocols with the highest iodine CNR value.
Not Applicable / None Entered.
Not Applicable / None Entered.