Automated pulmonary nodule tracking – development and evaluation
Principal Investigator
Name
Yiting Xie
Degrees
Ph.D.
Institution
Merative
Position Title
Data Scientist
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-949
Initial CDAS Request Approval
Aug 29, 2022
Title
Automated pulmonary nodule tracking – development and evaluation
Summary
Clinical management of CT detected pulmonary nodules often involves the monitoring and tracking of nodules over a period to assess the stability/changes of these nodules. Manually tracking and assessing pulmonary nodules can take up a significant amount of time from a radiologist. This project aims to develop automated methods to find and track pulmonary nodules over time.
Aims
In this project, an automated pipeline of computer algorithms will be designed and implemented to find and match pulmonary nodules over time in CT images. A subset of the image data will be used to train a computer model to track nodules between a prior CT scan and its corresponding current CT scan. Once the pipeline is developed, it will be evaluated on another subset of data to assess its accuracy (i.e. percentage of correctly tracked nodules). Furthermore, the efficacy of the automated pipeline in a real-world workflow will also be evaluated.
Collaborators
David Gruen, Merative, Jefferson Radiology (Hartford, CT)
Ehsanul Haque, Merative
Harald Zachmann, Merative
Marwan Sati, Merative