Predicting the post-test probability of lung cancer to reduce the time to the diagnosis
The objective of my research is to reduce the time to the diagnosis of lung cancer by analyzing the CT images of Indeterminate pulmonary nodules (IPN) for predicting the post-test probability of lung cancer at various intervals. Appropriate CT images are selected based on the inclusion criteria and then transfer to the server for de-identification and subsequent uploading in different destination softwares Healthmyne, Optellum, and Canary. This is followed by manual segmentation of the IPN and analysis of the same and using the results to use image biomarker for early detection of lung cancer and reducing the time to diagnosis. Clinical data elements are also being collected in collaboration with the University of South Carolina to build a bronchoscopy score calculator.