Clinical validation of a web- and cloud-based lung computer aided detection system

The purpose of our work is the clinical validation of a Computer Aided Detection (CAD) system for the automatic identification of pulmonary nodules in chest Computed Tomography (CT) scans. Non-calcified pulmonary nodules are the early manifestation of lung cancers. Lung cancer is the leading cause of cancer-related death worldwide. Screening high risk individuals for lung cancer with low-dose CT scans is now being implemented in the United States and other countries are expected to follow soon. The detection of these pathological Regions Of Interest (ROIs) is a burden task for radiologists, mainly due to the high number of noisy images to be analysed. To support radiologists, researchers have started implementing CAD algorithms for the automatic identification of pathological ROIs. Several studies proved the positive impact of CADs as a support for radiologists in the detection, with sensitively benefit on the overall performance. Despite these very prominent results, CAD systems have not been spread in clinical routine yet. In fact, the standard approach to make CAD algorithms available in the clinical routine of health facilities, that is the deployment of standalone workstations, usually equipped with a vendor-dependent Graphic User Interface (GUI), presents several drawbacks, such as the high fixed cost of the software licenses and the dedicated hardware and the rapid obsolescence of both. Furthermore, the computational needs by CAD algorithms can be demanding, depending on their complexity, often requiring powerful and expensive hardware. The diffusion of Cloud Computing solutions, accessible via secure Web protocols, solves almost all the previous two issues. In addition, the Software as A Service (SaaS) approach provides the possibility of combining several CADs, with demonstrated benefits to the overall performance