Alumni and Previous Staff

Contact

"This email address is being protected from spambots. You need JavaScript enabled to view it.
Download information as: vCard

Miscellaneous Information

Research Interest

Machine Learning, Pattern Recognition, Image Registration, Image Segmentation, Head and Neck Radiotherapy

 

About Me

I joined the PhD program in 2010 after working in various research facilities in the UK and North America. I completed an MSc in Computer Science in France and an MSc in Neuroscience in the UK. My research focuses on applying machine learning and data mining techniques for the development of atlas-based algorithms to obtain automatic segmentation of target volumes and organs at risk in the head and neck region. The acquisition of large imaging data prior to, during and after treatment may be a significant challenge for hospital logistics but offers a great opportunity for large scale analysis and data mining.

 

Research Description

Radiation Therapy (RT) requires the segmentation of organs at risk on various imaging modalities in order to maximize the dose received by the target tumor while controlling the dose received by the surroundings organs at risk. Manual contouring can provide these delineations but is dramatically time consuming and prone to inter-expert variability. My research focuses on applying machine learning techniques for the development of novel atlas-based algorithms to obtain automatic segmentation of organs at risk in the head/neck region. The aim is to develop algorithms that produce fast, accurate and consistent results that can be used for RT planning. The acquisition of large quantities of imaging studies prior to, during and after treatment may be a significant challenge for hospital logistics but offers a great opportunity for large scale analysis and data mining.

variability

Fig. 1: Significant variations between head and neck CT images render automatic segmentation to be an extremely challenging task. Some factors of variability: patient corpulence (I)-(II), head flexion (I)-(III), field of view (III)-(IV).

ct seg

Fig. 2: Automatic segmentation of the brainstem and spinal canal using the STEPS algorithm (Cardoso, 2013).

 

Publications

1. Hoang Duc A.K., Modat M., Leung K.K., Cardoso M.J., Barnes J., Kadir T., and Ourselin S.: Using Manifold Learning for Atlas Selection in Multi-Atlas Segmentation. (2013). PLoS ONE 8(8).

2. Hoang Duc A.K., Modat M., Leung K.K., Kadir T., and Ourselin S.: Manifold Learning for Atlas Selection in Multi-Atlas Based Segmentation of Hippocampus. (2012). SPIE.