Identification of valid, novel, potentially useful, and ultimately understandable pathologic patterns in large-scale medical image archives for computer-aided diagnosis
Professor David Feng, Dr Yong Xia.
Medical imaging has become a major tool in clinical practice, such as diagnosis and prognosis, since it enables rapid, non-invasive and in vivo visualization and quantitative assessment of the human body. Currently, the ever-increasing amounts of medical images have been produced in medical centers around the world. For example, the Royal Prince Alfred Hospital (RPAH, Sydney) houses one of the largest PET-CT image archives in the world, with more than 10,000 images, a number that is increasing daily. In this archive, each image has been inspected by medical professionals, and been attached with a detailed description and diagnostic report. Therefore, this archive implies a lot of valuable knowledge about diseases’ progression and responses to treatments, and precious experience of diagnosis. This project aims to discover the pathologic and diagnostic knowledge embedded in large-scale medical image archives by exploiting the joint-disciplinary medical image mining, which incorporates data mining techniques into medical image analysis and may serve as the basis for intelligent computer-aided diagnosis.
The opportunity ID for this research opportunity is 1194