Machine Learning to identify airway disease
Principal Investigator
Name
Amir Sharafkhaneh
Degrees
M.D., Ph.D.
Institution
Baylor College of Medicine
Position Title
Professor
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-921
Initial CDAS Request Approval
Jun 13, 2022
Title
Machine Learning to identify airway disease
Summary
Lung disease may stem from involvement of airways and/or lung parenchyma. Airway disease usually presents anatomically as thickening of small and large airways and presence of excess mucus in small and large airways. Parenchymal disease presents as destruction of airspaces, scaring and distortion of airspaces, inflammation and filling of airspaces or a combination of the aforementioned. In recent years, researchers use various large imaging databases in combination of various machine learning methods to identify above patterns of lung disease and correlate them with function and clinical outcomes. Current database may provide a better set of data for improving on current machine learning techniques to create an multidimentional index of lung anatomicopathological involvement.
Aims
1- Experiment on various machine learning methods to identify type and extend of damage to airways and/or parenchyma
2- Create a multidimensional lung health index that will reflect on the extent of lung involvement in part of the dataset
3- Validate such a multidimensional index in the rest of the data from this dataset
Collaborators
1- Jafar Tanha, PhD, Professor, Tabriz University
2- Alireza Safavi, Ph.D, Professor, Shiraz University