Development and validation of a computer-aided diagnosis system for lung cancer screening
Lung cancer screening in high-risk groups using LDCT has become very important worldwide after NLST reports that lung cancer mortality rates can be significantly reduced compared to screening using X-rays.
However, the number of radiologists for quantitative pulmonary nodule evaluation in lung cancer screening CT scans is not enough in the medical field, and there is a possibility that a radiologist may miss a nodule and a quantitative evaluation result may differ from person to person.
For accurate detection and quantitative evaluation of pulmonary nodules in LDCT and to reduce the workload of radiologists, artificial intelligence algorithms are currently being developed and are expected to be widely used in the medical field through the development and validation of algorithms.
In this project, we are trying to develop a deep learning-based algorithm for pulmonary nodule detection and quantification and we will validate its effectiveness through external validation by using the NLST datasets.
Our final goal is to improve diagnostic performance and clinical workflow through algorithm development and validation
- Development of a deep learning-based algorithm for lung nodule detection and quantification
- External validation of the internally evaluated system
- improving diagnostic performance and clinical workflow