Click here to

Session: Imaging: CT Image Quality, Protocol Optimization, and Dose Reduction [Return to Session]

A Method to Estimate the Half-Value Layer in CT From Two Exposure Measurements

R Al-Senan*, K Brown, S King, Penn State University, Hershey, PA


TH-IePD-TRACK 1-1 (Thursday, 7/29/2021) 3:00 PM - 3:30 PM [Eastern Time (GMT-4)]

Purpose: Among other factors, dose to a CTDI phantom depends on the beam quality. If contributions from scatter to exposure to center-hole is excluded one can, in theory, estimate the central beam half-value layer (HVL). The proposed study investigates the validity of this concept.

Methods: Transmission factors (TFs) through 15.4-cm PMMA were determined using a spectrum generator of beams whose qualities were modified by adding various aluminum filter thicknesses. The resulted HVLs and TFs from the simulations, at four tube-potentials, were fitted together using a second-order polynomial. To find the HVL from CTDI measurements, scatter-to-primary ratio (SPR) at each kV was determined. This was achieved through measurements in phantom (with 32-cm CTDI phantom, center-hole) and in-air on five scanners with HVLs manually-measured with Al filters. SPR was estimated as: [Center hole exposure / Primary exposure-1], where primary was calculated as: TF x in-air exposure, where TF was determined using the polynomial relationship. Finally, by estimating the TF from the mean SPR at each kV, HVL was calculated for every scanner and kV combinations and compared to those manually-measured. Also, manually-measured HVLs were compared to values measured automatically by a calibrated multi-sensor.

Results: SPR per kV did not vary significantly among different scanners/bowtie filters with different HVLs. Collimation showed no effect on SPR. Average SPRs were: 8.1, 7.6, 7.1 and 6.9 for 80, 100, 120 and 140 kV, respectively. Error of calculated vs. manually-measured HVLs was, on average, 4.7% (±0.34 mm) over all four kV stations. Measured HVLs from the multi-sensor were found to result in an average error of 10.8% (0.75 mm).

Conclusion: The method of estimating the HVL from in-air and in-phantom measurements was found to be reasonably accurate and practical. It can be used during the periodic CT testing. Furthermore, it showed higher accuracy than using the multi-sensor.



    Not Applicable / None Entered.


    Not Applicable / None Entered.

    Contact Email