Airway Inspector: a chest imaging biomarker software platform for COPD

Brigham And Women'S Hospital, Boston MA

Investigators

Linked publications, trials & patents

Abstract

DESCRIPTION (provided by applicant): Chronic Obstructive Pulmonary Disease affects up to 24 million people in the United States and is projected to be the 3rd leading cause of death worldwide by 2020. COPD has been traditionally dichotomized into the clinical phenotypes of emphysema and chronic bronchitis. Although spirometry is critical for diagnosis of COPD, FEV1 may be of limited value in a classification scheme designed to illuminate differences in disease mechanism among individuals. Computed tomographic (CT) imaging of the lung is a tool that may fulfill the need of new biomarkers to: facilitate the creation and testing of novel therapies, act as surrogate endpoints for clinical trials and provide researchers with the ability to reduce the heterogeneity of a cohort of subjects with COPD into more homogeneous subgroups to assist in monitoring study endpoints Airway Inspector is a open source software tool based on Slicer 2.x developed at Brigham and Women's Hospital for the analysis of chest CT scans for the analysis of emphysema and airway disease. The tool was conceived to analysis low resolution CT scans bringing the gap between retrospective data and new emerging CT technologies. The success of the tool in the COPD clinical community is supported by the array of peer- review publications that have used the tool to define metrics of disease that enable hypothesis driven research. Beside the existence of alternative commercial applications, Airway Inspector is the only freely available maintained platform for COPD research. In spite of its success, Airway Inspector is limited to a discontinued version of Slicer. The broad objective of this proposal is to support the refactoring and development of Airway Inspector as a platform for image-based COPD research. This goal will be achieved by the creation of the Chest Imaging Biomarker Platform (CIBP) library that integrates novel algorithm solutions to lung image analysis that have been developed in our laboratory. Those solutions include: robust lung extraction, parenchymal tissue classification based on local density, airway, fissure and pulmonary vascular extraction based on scale-space particles and airway labeling based on Hidden Markov Models. That software platform will be used to create workflows that will be integrated in Slicer 4 for their deployment in the clinical community. Slicer 4 is a NIH supported open software general purpose imaging software platform that is specifically designed to integrate solutions like Airway Inspector. The workflows will provide an end-to-end solution for the clinical to obtain phenotypes to characterize emphysema, airway disease and pulmonary vascular remodeling.