Application of Image Processing Techniques in Molecular Imaging of Cancer

Cancer is one of the leading causes of death all over the world.The research of cancer has always been a major focus in the medical imaging field. Modern imaging technology such as molecular imaging has already been shown to be useful in enhancing cancer management, including early detection, more accurate diagnosis, better treatment planning, and treatment monitoring in an early stage. Molecular imaging enables in vivo visualization and measurement of biological process at the cellular and molecular level. It allows not only localization of tumor lesions but also visualization of the expression and activity of specific molecules, which have great influence on tumor behavior and response to treatment. Many different modalities, such as MRI, SPECT, and PET, have been developed and used for noninvasive molecular imaging and have played a critical role in clinical oncology. However, due to large intersubject variability and various parameters in molecular images, it is generally infeasible to derive a single analytic method or simple equations that can describe the targets such as lesions and anatomies in all the images. Hence, in order to facilitate further the application of molecular imaging in clinical oncology, image processing techniques have been widely applied to the detection of cancer, characterization and segmentation of tumor lesions, planning of cancer treatment, evaluation of the effectiveness of treatment, prognostication of cancer, and so forth. Therefore, advanced image processing techniques have become a major focus in molecular imaging research, so that we can make better use of the rich information in the molecular image data.