(Monday, 7/30/2018) 1:15 PM - 1:45 PM
Room: Exhibit Hall | Forum 2
Purpose: To develop a methodology to study changes in pulmonary vasculature following thoracic radiation therapy based on quantitative imaging radiomics.
Methods: We consider 37 radiomic features as potential biomarkers for changes in pulmonary vasculature following radiation therapy. When extracted from CT-segmented lungs, these features collectively characterize intensity variations and texture patterns of the lung tissue. All features are extracted from (1) the pre treatment planning CT image and (2) follow-up CT images. Delta radiomics are then defined based on changes in feature values after therapy relative to baseline.This methodology was tested on a patient who developed pulmonary hypertension as a normal tissue complication of radiotherapy. The lung tissue indicative of pulmonary hypertension was segmented on follow up CT imaging by an experienced physician. The follow-up CT was then registered to the pre-treatment planning CT. Quantitative delta-radiomic features were derived and analyzed.
Results: Preliminary results show distinct changes in several radiomic features between the pre and post treatment CT images. In general, features derived from the gray level co-occurrence matrix and gray level run length matrix displayed the greatest change. Long Run Low Gray Level Emphasis demonstrated the most change in absolute feature value following radiation therapy. This suggests that density and texture changes in pulmonary vasculature may be predictive of pulmonary hypertension following thoracic radiation therapy.
Conclusion: This pilot study demonstrates the feasibility of a radiomics approach to quantify changes in pulmonary vasculature following radiation therapy for thoracic cancers.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
