Development and evaluation of algorithms for the lung disease diagnostics/management systems
Principal Investigator
Name
Yoshiro Kitamura
Degrees
Ph.D.
Institution
FUJIFILM corporation
Position Title
Primary Research Scientist
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-622
Initial CDAS Request Approval
Dec 26, 2019
Title
Development and evaluation of algorithms for the lung disease diagnostics/management systems
Summary
The object of this project is to develop and evaluate the computer-aided detection and diagnosis systems for lung diseases. We have been developing multiple systems based on deep learning technology in order to support diagnostic imaging workflow of lung diseases e.g.: lung nodule detection, lung nodule characterization, interstitial lung disease classification, radiology report generation. These systems have been developed and validated in collaboration with limited number of hospitals. However, thorough evaluation on diverse dataset remains to be done. Based on the performance shown on the diverse dataset, the systems will be developed further to overcome the weaknesses. The dataset may be used for designing, training, improving, evaluating or validating the model, workflow, prototype or software in the development and/or the developed products, systems or services.
Aims
1. Evaluate the performance of the computer-aided systems for lung disease, such as nodule detection, measurement, characterization and tracing on baseline/follow-up LDCT pair.
2. Develop the computer-aided systems to overcome the weaknesses found in evaluation study.
Collaborators
Imaging Technology Center, Research & Development Management Headquarters, FUJIFILM corporation
Medical System Development Center, Research & Development Management Headquarters, FUJIFILM corporation
Medical Systems Business Division, FUJIFILM corporation