Prevent Cancer Foundation quantitative CT imaging workshop XVI: lung cancer, COPD and cardiovascular disease - on the cusp of transformation, Arlington, VA

Abstract

Abstract Throughout the course of this Workshop, members of the advocacy community, imaging researchers, computer and measurement scientists, clinicians and policy-focused workshop attendees engaged in cross-cutting discussions from innovative technical aspects of thoracic imaging to policy approaches to ensure equitable access to all at-risk individuals when implementing lung cancer screening services. A major aspect of these implementation discussions was how to efficiently collect routine thoracic CT-based screening with de-identified clinical outcomes data to support the development of robust imaging tools, including responsible AI development, to better detect and manage early lung cancer as well as other major tobacco-related thoracic diseases. A future vision involves routinely collecting a substantial fraction of every thoracic screening CT image to establish a large, curated collection of de-identified thoracic CT images with clinical outcome data to support open research for building better computational imaging tools for early thoracic disease management. Imaging researchers are positioned to develop much better workflow software tools to promote more efficient, outpatient management of the screening process for populations at-risk for lung cancer, especially with the rapid development of promising AI tools. Efficient and effective management tools for the large numbers of at-risk ever smokers could allow the primary care community to discuss lung cancer screening despite their heavy existing clinical demands. Supporting the primary care community in this fashion may significantly improve the current slow uptake of lung cancer screening and save many lives in the process.

Keywords

• Lung cancer screening
• Tobacco-related diseases
• Quantitative imaging
• Artificial intelligence
• Image quality conformance measures
• Image archive