|
Abstract:
|
Radiomics is an emerging field of medical image analysis where radiological images are quantitatively analyzed to predict patient outcomes. For example, in cancer patients images of various modalities, including computed tomography (CT), magnetic resonance images, and positron emission tomography, have been used to quantify tumor heterogeneity and identify imaging predictors of patient survival. Rigorous data analytic techniques are required to estimate relevant image based measurements associated with disease progression as well as to develop prediction algorithms for outcomes of interest. In this talk, I will describe a data-driven approach that utilizes the existing methods of quantifying image-based information using machine learning techniques for outcome prediction.
|
