Lung Cancer, Chronic Obstructive Pulmonary Disease (COPD) and Cardiovascular disease (CVD) are called the Big-3 (B3) in health care. Together they cause the highest mortality, morbidity and economical strain on health care in the western world. The three diseases are mostly assessed using computed tomography scans (CT-scans). B3CARE aims to advance the technology readiness level for screening of the three diseases by accelerating software development for data post-processing of CT-images. To accomplish this, a robust B3 screening method is required as well as implementation of this method in the current health care.
To realise widespread implementation of B3 screening, it is necessary to inform/train (technical) physicians on how to exploit computer assistance based on post-processing software like machine learning and deep learning designed for assessment of these diseases. Better assessment of B3 disease can be achieved by integration of one’s own visual interpretation of image data with quantitative outcomes of machine intelligence. In this way indicating evidence on which a machine learning or software post-processing interpretation is based.
In the beginning stage of the project it will be assessed what the possibilities in the field of machine learning, and specifically deep learning, are in regards to the B3CARE project. Using this assessment it will become clear what subject(s) will be most interesting for application. This could consider techniques for visualising deep neural networks in assessment of the three diseases. Another example could concern the mimicking of CT-data, using deep learning, of the three diseases to create an endless database of CT-images. After an explorative phase, the assignment will focus on one of the mentioned examples. The eventual goal of the project will be to contribute to giving (technical) physicians insight in the use of machine learning and deep learning in the assessment of B3 diseases.