Lung nodule detection with artificial intelligence
Principal Investigator
Name
Anand Patel
Degrees
MD
Institution
Maverick Capital
Position Title
Consultant Researcher
Email
About this CDAS Project
Study
NLST
(Learn more about this study)
Project ID
NLST-101
Initial CDAS Request Approval
Dec 11, 2014
Title
Lung nodule detection with artificial intelligence
Summary
Smoking is a leading cancer risk factor, and specifically lung cancer is the leading cause of cancer mortality in the USA. Several attempts have been made in the past to perform computer aided detection of radiologic exams, including mammograms. Many of these applications have been limited by algorithm applications and lack of robust data sets. There have been more limited examples related to chest CTs. Over the last 10 years, artificial intelligence and computer vision has undergone a 'renaissance' with burgeoning technologies such as machine learning and neural networks. We aim to use state of the art machine learning and artificial intelligence algorithms to create a method for improved automated chest CT interpretation by radiologists. This dataset will be incredibly valuable in order to perform early stage testing of these algorithms. They will ideally lead to larger prospective multi-center randomized controlled trials of the algorithm.
Aims
1. Develop algorithm for lung CT interpretation.
2. Perform testing of algorithm to determine its accuracy.
3. Refine algorithm iteratively to develop more advanced methods.
Collaborators
Lawrence Livermore National Lab