(Tuesday, 7/31/2018) 10:00 AM - 10:30 AM
Room: Exhibit Hall | Forum 9
Purpose: Patients treated for non-small cell lung cancer (NSCLC) may suffer from comorbidities that weaken their functional condition. And yet, there are limited imaging techniques that can quantify the heterogeneous distribution of lung function prior to radiation therapy. A technique that has shown promise is a CT-based approach called topological parametric response mapping (tPRM). First demonstrated in COPD, tPRM generates spatially resolved maps of disease components, such as emphysema, functional small airway disease (fSAD), and parenchymal disease (e.g. ILD and pneumonitis). The purpose of this study is to explore the efficacy of tPRM to identify pulmonary dysfunction in NSCLC patients and improve radiation treatment planning.
Methods: High resolution inspiration and expiration CT scans were obtained from NSCLC patients through an ongoing IRB-approved clinical trial. Lung parenchyma are segmented on each scan and spatially aligned through deformable volumetric registration using a mutual information objective function. Topological analysis is performed to develop volumetric and surface area maps that represent the local accumulation and density of lung parenchyma with fSAD, fibrosis, or emphysema.
Results: tPRM analysis has been performed on 13 NSCLC patients. Of these patients, 8 were identified to have >5% accumulated volume of pulmonary dysfunction at pre-treatment. The voxelwise classifications of pulmonary dysfunction were combined with the delivered dose matrix to observe the effect of dose on changes in defected regions at mid- and post-treatment timepoints. Quantitative analysis of pulmonary dysfunction is being compared to qualitative scoring methods.
Conclusion: By implementing a tPRM analysis, pulmonary dysfunction can be spatially identified and the relative severity of each region can be quantitatively compared. Therefore, it is proposed that the tPRM method can be valuable in radiation treatment planning by identifying areas that are compromised by comorbidity, which can be subsequently accounted for in the radiation treatment design and analyzed for association with toxicity.
Funding Support, Disclosures, and Conflict of Interest: This research was partially supported through the P01-CA059872 research grant. Two of the authors (CJG, BAH) report a financial interest in the research through their affiliation with the company Imbio.
Keywords
Functional Imaging, Lung, Treatment Planning
Taxonomy
IM/TH- Image Analysis (Single modality or Multi-modality): Quantitative imaging
Contact Email
