Accelerating Learning Process for Medical Image Segmentation using Opposition-based Learning

In this project we will translate and test artificial intelligence techniques to segment medical images with the capability to perform more accurately than conventional methods. The technique is expected to largely eliminate tedious manual delineation of suspicious objects in medical images or, at least, significantly reduce the intensity and frequency of manual modifications traditionally preformed by radiologists. The intelligent segmentation will be observer-oriented and will include the radiologist's feedback into the computerized procedure of lesion delineation with the benefit of reducing necessary changes by radiologist. Consequently, the intelligent segmentation is expected to save time by automatically generating the outline of lesions/ organs/tissues in real-time. Over the duration of this project, this method will be applied to multiple modalitles and diagnostic cases.

Faculty Supervisor:

Dr. Allan Jepson

Student:

Pardis Beikzadeh

Partner:

OMISA

Discipline:

Computer science

Sector:

Life sciences

University:

University of Toronto