Author: Ryan Andosca, Igor Barjaktarevic, Peter Boyle, Jie Deng, Minji Victoria Kim, Michael Vincent Lauria, Daniel A. Low, Claudia R. Miller, Drew Moghanaki, Louise Naumann, Jack Neylon, Dylan P. O'Connell, Ricky R Savjani 👨🔬
Affiliation: Department of Pulmonology, University of California Los Angeles, University of California, Los Angeles, Department of Radiation Oncology, University of California, Los Angeles, UCLA, UCLA Radiation Oncology 🌍
Purpose: To develop a Hounsfield Unit ventilation-based correction method for use with model-based CT when used as a replacement for 4DCT.
Methods: The model-based CT we employ is termed 5DCT, which uses 25 fast-helical free-breathing (FHFBCT) CT scans along with a simultaneously measured breathing surrogate and a patient-specific motion model. This process generates CT scans at user-specified breathing states, but as of now only deforms a reference CT scan and does not alter the HU to reflect breathing ventilation. For each reference scan voxel, we used the voxel HU values of the 25 registered CT scans and their associated breathing amplitudes, with the slope of this relationship constituting the voxel's ventilation correction. We validated our correction by simulating 6MV transmitted fluences through the lungs, comparing 24 uncorrected and corrected FHFBCT scans across a group of 224 patients. To minimize the impact of stochastic noise, the HU value for each voxel was averaged from a 3-voxel cube centered on that voxel.
Results: The maximum uncorrected fluence error of a region between any two scans was found to be, on average across all patients, 2.78%. The same region and scan pair when corrected was found to have an average fluence error of 0.46%. The maximum corrected fluence error of a region between any two scans was found to be, on average, 1.16%. These significant reductions underscore the effectiveness of our correction approach.
Conclusion:
The fluence evaluation demonstrated that our approach to correcting the HU values output by our MBCT approach resulted in more accurate CT HUs, evidenced by substantially reduced transmitted fluence differences. These improvements are anticipated to enhance the precision of dose delivery in lung radiotherapy, potentially improving patient outcomes. Future work will focus on refining this correction technique and evaluating its potential impact in clinical settings.