Study
            
                PLCO
                (Learn more about this study)
            
 
            
            
                Project ID
                
                    
                        PLCOI-1509
                    
                
            
            
                Initial CDAS Request Approval
                Mar 25, 2024
            
            Title
            Pan-cancer analyses of imaging-clinical phenotype associations using updated PLCO cohorts
            
                Summary
                Machine learning methods can extract previously unknown associations from high-dimensional datasets. In this project, we will develop quantitative algorithms to process the radiology and histopathology images collected by the PLCO study. We will apply machine learning methods to identify the associations among radiology image patterns, microscopic pathology images, and clinical phenotypes of patients who participated in the updated PLCO cohorts. Our approach is expected to extract clinically relevant image patterns from radiology and pathology data.
            
            
                Aims
                Aim 1: To connect radiology image patterns with clinical phenotypes.
Aim 2: To associate the pathology image features with clinical phenotypes.
Aim 3: To correlate pathology and radiology image features.
Collaborators
                
                Kun-Hsing Yu, MD, PhD, Harvard Medical School.