Optimize radiological workflows with automated diagnostics
Principal Investigator
Name
Daniel Drieling
Degrees
M.Sc.
Institution
MeVis Medical Solutions AG
Position Title
Product Manager
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-491
Initial CDAS Request Approval
Mar 29, 2019
Title
Optimize radiological workflows with automated diagnostics
Summary
Lung cancer screening requires radiologists to read and report large amounts of imaging data. Computer-aided diagnosis software tools support the radiologist and reduce inter-observer variability to guarantee safety, effectiveness and reduce reading time for the benefit of a cost efficient high quality healthcare system. In order to create new or enhance existing software tools, a large amount of annotated images shall be used to provide clinically relevant results with respect to the detection and reporting of lung nodules. The NLST lung CT images shall be used for research and development to create and enhance software tools as well as to validate their clinical accuracy.
Aims
1) Develop computer-aided diagnosis (CAD) methods to detect and report lung cancer including nodule detection and malignancy probabilities.
2) Evaluate and validate the methods for clinical relevance, performance and effectiveness.
3) Benchmarking of specificity and sensitivity for existing CAD methods with variable sub-groups and cohorts.
Collaborators
N/A