Skip to Main Content

COVID-19 is an emerging, rapidly evolving situation.

What people with cancer should know:

Get the latest public health information from CDC:

Get the latest research information from NIH:

Principal Investigator
Peng Huang
Johns Hopkins University
Position Title
Associate Professor
About this CDAS Project
NLST (Learn more about this study)
Project ID
Initial CDAS Request Approval
Jun 17, 2016
Predicting overall survival using CT and pathological image features
We propose to extract texture features from CT and pathological images. The association between these two platforms of image markers will be investigated. Two models will be developed using machine learning for high dimensional data. One is the diagnostic model to predict time to cancer diagnosis using clinical and CT image markers; the other one is the overall survival model to predict time to death using clinical, CT image, and pathological image markers adjusting for treatments received.

Aim 1. To extract CT image texture features and to test the hypothesis that combining CT image texture features with clinical and epidemiological risk factors can reduce the false positive rate of CT image diagnosis.

Aim 2. To extract pathological image features and to identify markers from CT and pathological image features that are associated with overall survival.


Elliot Fishman
Edward Gabrielson
Junghoon Lee

Related Publications