Intelligent Multimodality Molecular Image Segmentation


Automated segmentation of major structures, tissues, and volume of interest (VOI) from PET/CT images new knowledge from information technologies.


Professor David Feng

Research Location

Computer Science

Program Type



he ever-increasing amounts of molecular images produced in hospitals around the world require intensive efforts of proficient physicians to manually delineate structures and region-of-interest (ROI) from them. This manual processing is laborious and operator dependent and thus prone to reproducibility errors. Automated image segmentation is traditionally a difficult task in image processing. However, the newly emerged multimodality molecular images have brought great potentials to the research of image segmentation, since they contain complementary information from diverse imaging modalities. The aim of this project is to use PET-CT images as our case study to investigate novel segmentation approaches for multimodality molecular images.

The output of this project will be a segmentation algorithm toolbox, which will facilitate molecular image visualization, retrieval, and computer aided diagnosis (CAD). Students involved in the projects will have the opportunity to work in RPA hospital for a certain period of time depending on the progress and necessarily. By finishing this project, students will learn some state of the art technologies in both medical imaging and multimedia.

Want to find out more?

Contact us to find out what’s involved in applying for a PhD. Domestic students and International students

Contact Research Expert to find out more about participating in this opportunity.

Browse for other opportunities within the Computer Science .


Life and medical sciences, Image segmentation, Image understanding, Image processing, Multi-dimensional data processing, Medical Imaging, PET/CT

Opportunity ID

The opportunity ID for this research opportunity is: 385

Other opportunities with Professor David Feng