Automatic Spondylolisthesis Grading

Approximately 5% of the population has a Spondylolisthesis (one of the lower vertebrae slips forward onto the bone directly beneath it). Non-invasive image-based assessment of this spinal structure is an increasing demand in healthcare service. However, the massive information embedded in the increasing large imaging scans has made manual inspection a lengthy and tedious task and observer dependent. The proposed research aims at a computer-aided spinal-scan assessment software toolkit facilitating the diagnosis of Spondylolisthesis. In particular, this proposal will develop and compare traditional (graph cut), machine learning, and recent deep learning methods in the diagnosis of spondylodesis, select in an efficient, reproducible and accurate method. This project is a pioneering attempt. In both industry and academia, there is limited prior work. Pros and cons of methods will provide a strategic base towards a full automatic spinal diagnosis system for the partner organization.

Intern: 
Dong Zhang
Chenchu Xu
Faculty Supervisor: 
Shuo Li
Province: 
Ontario
University: 
Partner University: 
Program: