Automated Assessment of Interstitial Lung Abnormalities
Principal Investigator
Name
Shikha Chaganti
Degrees
M.S., Ph.D.
Institution
Siemens Healthineers
Position Title
Research Scientist
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-616
Initial CDAS Request Approval
Dec 13, 2019
Title
Automated Assessment of Interstitial Lung Abnormalities
Summary
In our previous work, we developed an algorithm for detection of emphysema in lung CT scans using modern machine learning methods. We want to explore this work further to develop image-processing algorithms based on CT imaging to detect and quantify the extent of fibrotic and non-fibrotic interstitial lung abnormality (ILA) patterns such as ground-glass opacity, consolidation, mosaic attenuation, honeycombing, and reticular pattern. We have existing technology to segment lung lobes in CT scans. Based on this technology, we will perform segmentation of the lung parenchyma to calculate quantitative measures such as lung and lobe volumes and intensities. Based on these measures, and the presence and extent of lung abnormalities, we will perform an analysis to predict the prognosis in patients with ILA.
Aims
Specific Aims
1. Develop an automated detection algorithm to detect interstitial abnormalities.
2. Perform lung and lobe segmentation and compute quantitative measures.
3. Perform predictive analysis to predict prognosis of ILA, i.e., risk of developing lung cancer and risk of mortality.
Collaborators
Shikha Chaganti, Siemens Healthineers
Sasa Grbic, Siemens Healthineers