ABSTRACT

This chapter deals with detection and segmentation of the lung nodule and tumor, which can be used in various clinical applications, such as therapy response assessment. Therapy response assessment is critical for cancer patient management and new drug approval. Traditional methods to assess the response are based on measuring nodule and tumor size changes in one or two dimensions on computed tomography (CT) before and after therapy and can be biased. To investigate if changes in nodule and tumor volume can better assess therapy response, there is an urgent need to develop accurate and reproducible computeraided tools. Automatic detection and segmentation of lung nodule and tumor is a difŽcult task, as lung nodule and tumor often have various sizes and irregular shapes and they can grow closer or attached to surrounding structures of similar density and intensity.