Medical Imaging
With regard to medical imaging research, the current program focuses are on image processing, pattern recognition, computer vision, computer graphics, computer projections and data processing, as well as other related technological applications in medical and biological sciences. The objective is to help medical physicians enhance their diagnostic accuracy, or enable biologists to perform comprehensive analyses on complex biological mechanisms.
From a diagnostic perspective, each kind of medical imaging has its own significance, with one complementing the other, offering a comprehensive picture of a particular disease. Therefore, researchers may miss important information by concentrating on one type of imaging systems. As such, it is critical to make correlations among a variety of medical images, auto-integrate diagnostic information from these images, and develop more effective ways to diagnose diseases through the combined use of different medical imaging systems. Examples of such studies include computer-aided nuclear medical imaging detection and diagnostic systems of the heart, 3D reconstruction of breast magnetic resonance images and registration of multimodal breast images, model-based 3D reconstruction and registration of microcalcifications in mammography images, as well as the retrieval of compressed and non-compressed images, etc.
Side view (4- degree angle) and front view of a 3D reconstruction of a breast
model
For the dissection of biological mechanisms, 2D gel electrophoresis has been widely applied to identify the tens of thousands of proteins in a cell, their expression and modifications. Hence, it is important to develop and establish tools for the analysis of large-scale proteomics data. Examples of such studies include the detection, comparisons, and prediction of protein spots in a 2D gel electrophoresis image, as well as the integration of 2D electrophoresis images with genome and proteome databases.
Identification of protein spots in a 2D gel electrophoresis image
Comparisons of protein spots in a 2D gel electrophoresis image